Sage Journals: Discover world-class research

Abstract

Against the background of tourism recovery following the COVID-19 pandemic, this study explores how destination personality shapes tourists’ emotions and their behavioral intentions at heritage sites. A framework is constructed to evaluate these three variables based on the Cognitive-Affective-Conative (CAC) theory. The methodology combines user-generated content (UGC), sentiment analysis, geographic information systems (GIS), and structural equation modeling (SEM). Online reviews serve as the data source for assessing both perceived destination personality and tourists’ emotional experiences. The statistical findings demonstrate that the personality of a place significantly impacts the emotions experienced by tourists. Among personality dimensions, prominence and responsibility most strongly affect tourists’ behavioral intentions. The research provides valuable insights for tourism managers and policymakers with guidance for fostering sustainable and responsible tourism development.

Plain Language Summary

A Study on Place Character, Tourists’ Emotions, and Intentions

In the context of tourism recovery, the character of a place is an important factor in attracting tourists. However, previous research has ignored its relationship with visitors’ emotions and intentions. This research uses online reviews with place analysis and statistics to reveal how the destination influences visitors’ emotions and intentions. The data were collected from Google reviews to measure the places and the emotional experiences. The findings show that the character of a place has an impact on the emotions experienced by tourists. “Well-known” and “well-cared” are the important characters. Thus, heritage managers can use this low-cost method to see how to keep the sites popular.

Keywords

destination personality tourist emotions sentiment analysis spatial analysis sustainable tourism

Introduction

The tourism industry has experienced substantial changes since the COVID-19 pandemic. Digital transformations have been accelerated, such as contactless payment and virtual traveling. Global competition is intensifying, driven by increasingly diverse traveler preferences, including nature-based, low-carbon, and responsible tourism (Adongo et al., 2024). In response, creating a good tourist destination image (TDI) in marketing, namely the impressions that a person has of a place (Gunn, 1972), has become essential for post-pandemic recovery, particularly in Southeast Asian economies that rely heavily on tourism. Based on the concept of TDI, different cities and destinations have built their own characteristics, which are referred to as their “personalities” that differentiate them from competitors. For example, Paris is associated with a strong brand related to romance or luxury (Dupain & Novitskaya, 2015); tranquility has emerged as a key personality trait attracting visitors elsewhere (Quintal et al., 2024). However, there are some problematic phenomena that have been observed: many destinations now project increasingly similar images. Misleading or similar promotional content may weaken consumers’ trust and attitudes toward these destinations (X. Li et al., 2020).

Many countries and cities are striving to create distinct brands in an increasingly competitive market. Their goals include not only differentiating the brand positioning of tourism cities but also deepening the emotional connections between people and touristic places. Destination management organizations (DMOs) need to understand tourists’ specific preferences, identify which personalities are most recognizable, determine how these traits shape perceptions, and evaluate their behavioral consequences. Researching these connections and identifying key factors that influence tourists’ behaviors can help enhance the perceived value of the destinations, encourage repeat visits, and stimulate positive word-of-mouth (WoM; Nella, 2022).

Especially nowadays, UGC from popular social media platforms, widely used for sharing experiences and promotional information, has been emphasized as a significant lens through which to explore the perceptions and feelings of tourists (Ha et al., 2023; Kar, 2021). With the evolution of the internet and communication methods since the 21st century, many organizations have begun using social media as a branding tool (Assaker, 2020; Ha et al., 2023). For example, the Tourism Authority of Thailand has launched dedicated Instagram accounts for this purpose. Such content can be interpreted as business intelligence for improving marketing decisions and maintaining competitiveness, particularly when promoted by official organizations or local agencies. UGCs contain various formats, such as ratings, complaints, and blogs (Lee et al., 2020; C. Liu et al., 2019). Researchers have tried to utilize them to develop predictive models, inform policy, and address destination-level challenges (Y. Zhang et al., 2024). Specifically, reviews (like Google Maps) capture consumers’ choices, trust, satisfaction, and attitudes, and have become a focal point of recent research (Mathur et al., 2022; Zervas et al., 2021). By analyzing this published content (Dabbous & Barakat, 2020; Du et al., 2022), site branding and experience enhancement can be achieved. Previous literature has also highlighted the potential use of reviews to examine tourists’ emotions, destination image, and behavioral intentions (Su et al., 2022; L. T. T. Tran, 2020; Yu & Zhang, 2020). And UGC emotions also play a crucial role in shaping tourists’ experiences and significantly influence users’ cognitive evaluations, behavioral responses, and decision-making (Hosany et al., 2020). Statista (2014) reported that tourists experience a range of emotions (excitement, control, knowledge, etc.) during their visits. Some real-time emotion sensors are invented to serve health/service sectors (Saadon et al., 2023; K. J. Wang et al., 2023). However, previous research has often overlooked the relationship among destination personality, tourist emotions, and behavioral intentions, particularly from a spatial perspective (Ekinci & Hosany, 2006; Gnoth et al., 2007; Hosany et al., 2007; Usakli & Baloglu, 2011).

Thus, this study tries to explore the correlations among destination personality, emotion, and behavioral intentions. It aims to address the following questions: (1) What are the relationships between destination personality, tourists’ emotions, and intentions? (2) How can these constructs be quantified and mapped using spatial methods? In an innovative approach, it employs several open-source tools for the quantification: sentiment analysis and machine learning by Orange, spatial interpretation via QGIS, and statistical methods by Statistical Product and Service Software Automatically—SPSSAU. It seeks to recognize specific feelings expressed and perceived personality of various places using lexicon-based data, and to predict emotional and personality patterns in areas surrounding key tourist sites. The significance of this research is multifaceted: a well-defined destination personality converts unique local attributes into competitive assets, which strengthen brand preference and increase revisit and repurchase intentions (X. Li et al., 2020). The findings also provide insights for promoting sustainable development in heritage cities and comparable tourism destinations.

Literature Review

Theoretical Foundation

“Brand personality” refers to a series of personified characteristics attached to a product or brand. Aaker (1997) defined the Brand Personality Scale (BPS), which includes five dimensions: sincerity, competence, sophistication, excitement, and ruggedness. Building on this framework, scholars have extended the concept to destination personality (Gunn, 1972; Hunt, 1975; Telisman-Kosuta, 1989): human personality traits associated with tourist places. Subsequent studies have applied the BPS to assess the personality traits of destinations and confirmed several key factors: excitement, sincerity, and conviviality (Hosany et al., 2007). Pan et al. (2017) pointed out that the original BPS neglected intangible place-specific qualities. In response, researchers have adapted or shortened the scale to fit various cultural and contextual settings (e.g., Davari & Jang, 2024; Jovanović et al., 2017; Kovacic et al., 2020; Kumar & Nayak, 2018; Papadimitriou et al., 2015; C. Zhang et al., 2019). Recently, a five-dimension scale was developed for World Heritage Sites (WHS): responsibility, exceptionality, prominence, identification, and attractiveness. It offers flexibility through the use of pre-trained language models and structured dictionaries (Hassan et al., 2024).

From a theoretical perspective, the Cognitive-Affective-Conative (CAC) theory is a psychological model that posits individuals’ intentions are shaped by cognitive beliefs and affective responses. In tourism, this theory has been widely used to provide insights into the complex interaction among affection, behavior, and cognition in shaping tourists’ experiences and behaviors. Recently, Tang et al. (2022) integrated the CAC theory with the Motivation-Opportunity-Ability model to explore the mechanism of environmental behaviors, confirming significant influences of motivation, knowledge, and opportunity. Yang et al. (2022) merged individualism and uncertainty avoidance into the CAC model and suggested that tourists’ affection mediated the relationship between conative and cognitive destination image. The application and development of this theory in tourism research are still under discussion, and more empirical studies are needed for further validation. Emerging studies increasingly integrate advanced tools: machine learning and computer graphics, to test the model (Alzboun et al., 2023).

Among alternative theories, Stimulus-Organism-Response is a micro-level theory in which environmental factors trigger internal states that produce approach or avoidance results. It focuses on sensory-driven responses, but less applicable to destination branding. Place Attachment Theory centers on a long-term emotional bond formed through across repeated visits, powerfully predicting loyalty. But this attachment is difficult to extract from UGC for rapid marketing adaptation. The Theory of Planned Behavior evaluates attitude toward the act, subjective norms, and perceived behavioral control. But these constructs have limited use for sentiment-based spatial analytics. Consequently, CAC offers the pathway for linking spatially sentiment to destination personality and, ultimately, to the revisit intentions that drive sustainable heritage-tourism strategies. Therefore, the theoretical foundation of this study is mainly based on CAC theory to examine the relationships among destination personality, emotion, and conation.

Hypothesis Development

Visitors’ emotional experiences are increasingly recognized as an indispensable part of branding (X. Liu & Zheng, 2024; Soon et al., 2023). The characteristic of a place was proven to be useful to increase consumers’ loyalty (Ghorbanzadeh & Rahehagh, 2021; Sung & Kim, 2010; Thomson et al., 2005). Destinations evoke emotions through attributes such as novelty, meaningfulness, cultural richness, and opportunities for learning (Hosany & Gilbert, 2010; Kim et al., 2012; Şahin & Güzel, 2020). Several studies proved a positive relationship between tourists’ emotions and the perceived destination images (Hosany et al., 2007; Kastenholz et al., 2020; San Martín & Rodríguez del Bosque, 2008). Tourists’ emotional experiences and destination personality also influenced each other. For example, Kovacic et al. (2020) found that the affective image of a destination is closely associated with its personality, particularly the traits of sincerity and excitement. Thus, to further confirm the relationship with the dimensions of responsibility, identification, exceptionality, prominence, and attractiveness (Hassan et al., 2024), this research hypothesizes:

H1: Destination personalities significantly affect tourists’ emotions.

The positive influences of destination personality on tourists’ intentions have been widely documented (H. Chen et al., 2024; Soltani et al., 2021; J. Wang et al., 2022; Yang et al., 2020). Suryaningsih et al. (2020) used Moderated Structural Equation Modeling (MSEM) to test the relationship between personality of destinations, self-image congruity, and behavioral intention. It showed that satisfaction and certain personality features can positively predict behavioral intentions in Serbia. Kovačić et al. (2022) examined the interplay between tourists’ individual traits, destination personality, and activity preferences, concluding that sincerity and ruggedness significantly shape tourists’ choices through their activity patterns. Based on a literature review, this research will test how personalities influence tourists’ intentions and proposes the following hypothesis:

H2: Destination personalities significantly affect tourists’ behavioral intentions.

Tourists’ intentions represent favorable or unfavorable willingness, including satisfaction, repurchase, loyalty, complaints, etc. Emotions can lead to specific responses in tourism. Different emotions may affect the decision-making process during the planning stage and leisure activities (Agyeiwaah et al., 2021). While on site, tourists can shift from frustration to delight depending on the situation (Göker & Çelik, 2022; Johnson et al., 2023). And these moment-to-moment feelings shape how memorable the experience is and whether intentions to revisit or recommend are formed (Hosany & Prayag, 2013; Mcintosh & Siggs, 2005). After the tour, positive emotions heighten tourists’ propensity to recommend the destination, return, or remain loyal (Bigné et al., 2005; Huang et al., 2012; H. Li et al., 2022). Conversely, negative emotions can breed risk perceptions; their frequency, timing, and intensity can deter future behavior (Torres-Naranjo et al., 2021). For example, Volo (2021) stated that fear can significantly influence tourists’ decision-making, leading to avoidance of certain destinations that are perceived as risky, while surprise can be double-edged: positive surprise enhances satisfaction, but negative surprise can reduce revisit intention (Neidhardt et al., 2015). Based on these findings, the hypothesis is proposed as two parts:

H3: Tourists’ positive emotions significantly affect their intentions.

H4: Tourists’ negative emotions significantly affect their intentions.

Tourists’ emotions may mediate the traveling process. Yao (2013) demonstrated that tourists’ emotional involvement mediates the effect of motivation on satisfaction, indicating that the affective dimension of a destination operates as a mediator. Similar mediation pathways between cognition and behavior have been confirmed in health and psychological research (Kiviniemi et al., 2018). Within the tourism context, the emotional mechanisms linking destination personality to intention remain under-explored. Thus, following the CAC framework, this study further proposed the hypothesis (Figure 1):

H5: Tourists’ positive/ negative emotions mediate the relationship between destination personality and tourists’ intentions.

Figure 1.

Variables in the framework.

Methodology

Sentiment analysis has been increasingly applied in the tourism sector, drawing on data from social media, blogs, and news outlets. It supports destination management by informing market segmentation, brand positioning, and growth strategies. Nevertheless, extant approaches have been criticized for questionable discriminant validity, omitted dimensions, inconsistent operationalization, and debatable covariates. And the application of machine learning and sentiment analysis in review collection presents several advantages over traditional methods, like questionnaires or interviews, which seldom capture the subtleties of natural language due to the complexity of language. Machine learning models trained for sentiment can understand context and misapplied words, providing a better understanding of customer sentiment. They can lower cost, accelerate data collection, and allow managers to react rapidly to customer feedback (Agarwal et al., 2016; Mitra & Mohanty, 2020). As large-scale UGCs have been considered a significant element in conceptualizing smart tourism, they have immense potential for increasing opportunities by combining with text-mining algorithms to produce accurate and flexible predictions of tourist behavior (N. Liu & Shen, 2020).

Based on the literature review, the research utilized a comprehensive approach involving machine learning, natural language processing, and structured dictionaries and rules to examine online reviews with George Town in Malaysia as a case study area. To quantify the three variables, the research methodology mainly followed previous research (Demšar et al., 2013; Hassan et al., 2024; Mehra, 2023). The main process is: data collection and cleaning, sentimental analysis, destination personality analysis by dictionary, tourists’ intention mapping by inverse distance weighted (IDW) method, and covariance-based structural equation modeling (CB-SEM; Figure 2). Statistically, SEM is widely used for testing measurement and hypotheses, specifying and estimating relationships, and analyzing complicated effects (Darda & Bhuiyan, 2022). Especially, CB-SEM can be used to test the fit of a theoretical model, whether the relationships among variables align with the model. The constructs in this research are reflective, namely, a latent factor structure with reflective indicators (Freeze & Raschke, 2007). The CB-SEM was selected for the confirmatory purpose of the model and to test the correlations among indicators (Hair et al., 2017).

Figure 2.

Workflow.

Study Area

George Town is a famous destination in West Malaysia, renowned for its well-preserved colonial and multicultural architectural heritage. The city has been experiencing a boom in tourism since 2008 (UNESCO, 2008), which generates both benefits and burdens (Pei & Nor Azila, 2020). For example, tourism has brought some economic benefits to George Town by creating job opportunities and improving infrastructure and amenities for tourists. But several challenges have also been presented, such as overcrowding, transportation and living pressure, damages to heritages, etc. (Ghaderi et al., 2012). The influx of outsiders has intensified debates over cultural authenticity, resident displacement, and the commodification of historic properties. Municipal authorities now prioritize responsible tourism, heritage-sensitive conservation, and participatory planning that engages local communities (Mohaidin et al., 2017). Because of these competing pressures and highly salient heritage landscape, George Town offers an ideal setting in which to examine how tourists’ emotional experiences and place personality construct a heritage destination (Figure 3).

Figure 3.

Location of George Town and selected samples.

Data Collection and Cleaning

Google reviews served as the primary data source for this research. Previous studies have proved the value of analyzing them in marketing and tourism (Guo et al., 2021). The data collection process focused on several popular destinations in George Town. In total, 23 main sites were chosen according to purposive sampling (site popularity). The collection period was between January 2023 and May 2023, during which 14,919 reviews were collected using a Python-based scraper (https://github.com/georgekhananaev/google-reviews-scraper-pro). Data cleaning was performed using Orange, an open-source machine learning software. After preprocessing steps, including conversion, removal, tokenization, filtering, stemming, and lemmatization, 12,513 valid records remained, accounting for 83.8% of the original dataset.

Computing Sentiments

This research computed both the polarity and various emotions to obtain the intention of tourists. Review polarity was categorized as “Neutral, Negative, and Positive” according to Hu and Liu’s (2004) sentiment lexicon. Polarity score were calculated as the difference between the total counts of positive and negative words, normalized by review length. Each review was then assigned a polarity label: positive (P), negative (G), or neutral (N). Following Mehra (2023), emotional polarity (S1) was determined as follows:

\begin{matrix} \begin{matrix} If N > 0.6 amp P > G \to S = P + G; \\ Elseif N > 0.6 amp P < G \to S = P; \end{matrix} \\ Elseif N < 0.6 S = P + N * (1 - G) \end{matrix}

(1)

The emotion scores were then classified into six categories based on Ekman’s emotion wheel: anger, fear, joy, sadness, surprise, and disgust (Janocha et al., 2018). Quantification was performed with a Support Vector Machine (SVM) implemented in Orange’s Tweet Profiler widget. Orange is an open-source, Python-based platform for machine learning and interactive data visualization (Demšar et al., 2004; Dobesova, 2024). SVM offers the flexibility to extract precise features from any domain-specific corpus (Colneric & Demsar, 2020). Compared with bag-of-words or latent semantic indexing approaches, the trained Tweet Profiler has superior accuracy. For each selected site, average emotion scores were computed. Finally, a tourist-intention score was derived by summing emotional polarity (S1) and the surprise value; higher scores indicate a stronger propensity to revisit (Mehra, 2023).

Destination Personality

To quantify personality (PE), the research adopted the dictionaries compiled by Hassan et al. (2024). Each personality list contains seed words that capture the respective dimension: for example, “affected,”“protected,”“preserved,”“concerned,”“responsible,” and “controlled” signal responsibility, whereas “common,”“essential,”“valuable,” and “significant” indicate prominence. Lexicon matching was performed in Orange: the software counts the occurrences of dictionary words (n) and normalizes by review length (L; equation 2):

PE = n \div L

(2)

The resulting value reflects the intensity of each personality trait conveyed in tourists’ reviews. Site-level scores were obtained by averaging the five personality dimensions. With emotion, personality, and intention scores computed, their inter-relationships could then be examined. Normality was verified through skewness and kurtosis in Excel, both of which fell between −1 and +1, indicating an approximately normal distribution suitable for subsequent analysis (Hair, 2006).

Spatial Analysis

QGIS (Quantum GIS) was employed to visualize spatial patterns and trends in the computed emotion and personality scores. Unlike other commercial spatial software, QGIS is open-source and extensible through a large ecosystem of plugins (Moyroud & Portet, 2018; Rosas-Chavoya et al., 2022). Site-level averages were interpolated into a continuous raster surface, allowing the emotional and personality landscapes to be mapped across the study area. Inverse Distance Weighting (IDW) was selected for the interpolation: unknown values are predicted as a weighted average of the nearest sampled points, with weights decreasing as distance increases (Masoudi, 2021):

w_{i} = \frac{1}{d_{i}^{p}}

(3)

z (x, y) = \frac{\sum_{i = 1}^{N} w_{1} z_{i}}{\sum_{i = 1}^{N} w_{1}}

(4)

In equation 3, w_i means the i-th weight of the point, d_i means the distance from the unknown points to the known points, and p means the controllable parameter. In equation 4, z (x, y) means the location of the unknown point; N means the number of known points; the coordinate of the i-th point is (x_i, y); and z means the values obtained by inverse function.

Structural Equation Modeling

Before running CB-SEM, a confirmatory factor analysis (CFA) was conducted to verify the measurement model. All analyses were performed with SPSSAU. For discriminant validity, standardized factor loadings had to exceed 0.6, AVE ≥ 0.4, and CR ≥ 0.6 (Lam, 2012; Truong & McColl, 2011). Model fit was evaluated with GFI, RMSEA, RMR, CFI, and other indices. The structural model supplied path coefficients, significance tests, and overall fit statistics. Acceptable fit required SRMR < 0.10 and other indices (NFI, CFI, NNFI) >0.90 (Hooper et al., 2008). Finally, emotion was examined as a mediator of the personality - intention link. Following Preacher and Hayes (2008), the direct (c′), indirect (ab), and total (c) effects were estimated; significance was inferred when paths a (X-M), b (M-Y), and c (X-Y) all yielded p < .05.

Results

Emotion and Personality Distribution

The IDW was utilized to draw raster data for the whole study area. In QGIS, the pixel size was set as 10 m. The interpolating is shown in Figure 4a and b (six emotions, five personalities, and tourist intention). It can be observed that the emotions of surprise, sadness, and fear were mainly concentrated in the center and east parts of George Town, showing an increasing trend from east to west. For the whole city, the average values of surprise, sadness, disgust, and anger were relatively low, indicating that negative emotions were not strongly expressed. Joy was the emotion with a high average value (0.995). It showed a dispersed pattern, with lower values concentrated in the center and higher values in the northern and southern parts. This showed that the central areas, such as the Han Jiang Ancestral Temple (traditional Chinese religious site), were the destinations perceived negatively by tourists. The correlation analysis (Table 1) showed that the negative emotions of surprise, sadness, fear, disgust, and anger were consistent with each other (p value < .001). However, the joy emotion was correlated negatively to the other emotions, which showed that the positive and negative emotions were totally opposite to each other.

Figure 4.

(a) Spatial mapping for emotions. (b) Spatial mapping for the destination personalities.

Table 1.

Correlations.

Feature 1	Feature 2	Correlation	p-value	Feature 1	Feature 2	Correlation	p-value
Fear	Joy	−.999	<.001***	Fear	Surprise	.545	<.001***
Disgust	Sad	.898	<.001***	Disgust	Joy	−.376	<.001***
Sad	Surprise	.819	<.001***	Disgust	Fear	.374	<.001***
Anger	Sad	.805	<.001***	Prominence	Responsibility	.718	<.001***
Anger	Joy	−.769	<.001***	Attractiveness	Exceptionality	.672	<.001***
Anger	Fear	.768	<.001***	Exceptionality	Prominence	−.589	<.001***
Joy	Sad	−.61	<.001***	Exceptionality	Responsibility	−.452	<.001***
Fear	Sad	.608	<.001***	Attractiveness	Identification	.330	<.001***
Anger	Surprise	.600	<.001***	Identification	Prominence	.284	<.001***
Anger	Disgust	.581	<.001***	Identification	Responsibility	.211	.003
Disgust	Surprise	.556	<.001***	Exceptionality	Identification	−.113	.123
Joy	Surprise	−.547	<.001***	Attractiveness	Prominence	−.112	.128
				Attractiveness	Responsibility	−.072	.328

***

p < .001 (two-tailed).

For the personality values, the ranking was: attractiveness > exceptionality > prominence > identification > responsibility, which meant that attractiveness (0.802) was the most distinctive feature. The spatial pattern identified was that identification and attractiveness showed an increasing trend from south to north; exceptionality, prominence, and responsibility showed a pattern changing from east to west. For the intention values, the distribution was not clear. A correlation was run to check the potential relationship among them. Table 1 showed that prominence and responsibility were strongly correlated (0.718); attractiveness and exceptionality were strongly correlated (0.672); exceptionality and prominence were negatively correlated with each other (−0.589); exceptionality and responsibility were also negatively correlated (−0.452); and between attractiveness - identification and identification - prominence, the correlations were moderate (0.330, 0.284).

SEM and Mediation Test

According to Table 2 (CFA results), several indicators within the personality and emotion variables contributed little to the model (factor loading < 0.6) and were therefore removed: identification, attractiveness, exceptionality, surprise, and disgust. After consultation with relevant experts and scholars, the emotions were reclassified into two groups: positive and negative. Following the deletion of these indicators, the revised model demonstrated good quality, with composite reliability (CR) exceeding 0.5 and average variance extracted (AVE) above 0.5. The Heterotrait–Monotrait ratio (HTMT) was below 0.9 (Afthanorhan et al., 2021), and additional fit indices met acceptable thresholds: GFI > 0.9, RMSEA < 0.10, RMR < 0.05, and CFI > 0.9. The refined variables satisfied discriminant validity requirements and were retained for path and mediation analysis.

Table 2.

CFA Results.

Variables	Indexes	Std. factor loading	Std. error	z value	p-value	AVE	CR	HTMT
Original model
Personality	Prominence	.943	-	-	-	0.401	0.719	0.427
	Identification	.285	0.075	3.73	.00***
	Attractiveness	−.168	0.102	−2.193	.028**
	Exceptionality	−.631	0.109	−8.392	.00***
	Responsibility	.756	0.072	9.799	.00***
Emotion	Anger	.769	-	-	-	0.567	0.877
	Surprise	.548	0.12	7.841	.00***
	Sad	.610	0.085	8.85	.00***
	Fear	.999	0.059	16.414	.00***
	Joy	−1.00	0.06	−16.422	.00***
	Disgust	.376	0.075	5.246	.00***
Revised model (after adjustment)
Negative emotion	Anger	.852	-	-	-	0.719	0.885	0.784
	Fear	.926	10.763	0.095	.00***
	Sad	.792	9.4	0.118	.00***
Positive emotion	Joy	.769	-	-	-	0.592	0.592
Personality	Prominence	.631	-	-	-	0.615	0.758
Personality	Responsibility	.946	0.726	1.817	.00***	0.615	0.758
GFI = 0.986, RMSEA = 0.096, RMR = 0.003, CFI = 0.993

p < .05 (two-tailed). ***p < .001 (two-tailed).

Table 3 and Figure 5 confirm the relationships among the three variables; all indicators contribute significantly (p < .001). For the path personality - positive emotion (p = .221) and personality - negative emotion (p = .402), Hypothesis 1 is rejected. Personality - intention is significant (p = .012, coefficient = 0.235), so Hypothesis 2 is supported. Positive emotion - intention is significant (p < .000, coefficient = 0.052), supporting Path 3. Negative emotion - intention is significant (p < .000, coefficient = −0.784), supporting Path 4.

Table 3.

SEM Results.

Variables	Indexes	Non-std. coefficient	Std. coefficient	z value	S.E.	p-value
Personality	Prominence	1	0.597	-	-	-
Personality	Responsibility	1.472	1	2.677	0.55	.007**
Negative emotion	Anger	1	0.855	-	-	-
	Fear	1.102	0.918	11.613	0.087	.000***
	Sad	1.118	0.801	9.676	0.116	.000***
Positive emotion	Joy	1	0.855	-	-	-
Intention	Intention	1	0.888	-	-	-
GFI = 0.973, NFI = 0.932, RMR = 0.007, CFI = 0.948

p < .05 (two-tailed). ***p < .001 (two-tailed).

Figure 5.

Path test.

Then, Table 4 confirmed that negative emotion played a mediator role between personality and intention. Paths a (destination personality—negative emotion) and b (negative emotion—intention) were significant (p < .001), but path c’ (destination personality—intention) was not significant (p > .05). Bootstrapping, a procedure that resamples the data to simulate the sampling distribution, showed that the indirect path had a 95% CI of −0.285 to −0.043, which does not cross zero; thus, negative emotion fully mediates the relationship between destination personality and tourists’ intention. By contrast, the 95% CI for positive emotion ranged from −0.231 to 0.003, crossing zero. So, positive emotion did not mediate the process.

Table 4.

Mediation results.

Path 1	Abb.	Meaning	Coefficient	95% CI		SE	z	p	Effect
Path 1	Abb.	Meaning	Coefficient	Upper	Lower	SE	z	p	Effect
Personality => negative emotion => intention	a*b	Indirect	−.068	−0.183	−0.036	0.037	−1.834	.067	Full effect
Personality => negative emotion	a	X=>M	.113	0.047	0.179	0.034	3.35	.001***
Negative emotion => intention	b	M=>Y	−.601	−0.824	−0.377	0.114	−5.27	.00***
Personality => intention	c’	Direct	.082	−0.034	0.199	0.06	1.386	.167
Personality => intention	c	Total	−.018	−0.126	0.089	0.055	−.337	.737
Path 2	Abb.	Meaning	Coefficient	upper	lower	SE	z	p	Effect
Personality => positive emotion => intention	a*b	Indirect	−.114	−0.231	0.003	0.06	−1.908	.056	Not significant
Personality => positive emotion	a	X=>M	−.148	−0.295	0.000	0.075	−1.966	.049**
Positive emotion => intention	b	M=>Y	.772	0.582	0.963	0.097	7.959	<.001***
Personality => intention	c’	Direct	.599	0.403	0.796	0.10	5.968	<.001***
Personality => intention	c	Total	.485	0.26	0.711	0.115	4.217	<.001***

Note. Estimator is ML.

Multiply. **p < .05 (two-tailed). ***p < .001 (two-tailed).

Discussions

Variable Relationships

Destination Personality and Emotion

Based on the SEM results, destination personality has a weak influence (−0.08 to 0.1) on tourists’ emotions at George Town sites. While the CAC theory shows a sequence from cognition (personality perception) to emotion, this research suggests a link may be more complex and indirect in a heritage tourism context. This weak effect partially aligns with C. F. Chen and Phou (2013), but contrasts with studies that found stronger direct relationships (e.g., Hosany et al., 2007). This discrepancy may be attributed to the nature of UGC: emotions in reviews are triggered by specific and momentary experiences (e.g., a long queue, a friendly guide) that may not be directly attributed to the overarching destination personality. Furthermore, in a culturally rich setting like George Town, emotions might be more strongly influenced by direct sensory experiences (e.g., street art) than by other abstract personality dimensions (V. A. Tran et al., 2025). Accordingly, it is suggested that tourism cities like George Town should focus more on diverse strategies to create an emotional connection with tourists. This can be achieved by promoting distinctive cultural and historical aspects, improving service quality, and creating memorable experiences. It is still important for tourism cities to understand the diverse needs and preferences of different tourist groups and to tailor their strategies accordingly. For instance, introverted people might prefer calmer destinations, while extroverted tourists might enjoy places offering exciting experiences (Neidhardt et al., 2015; Scuttari et al., 2016).

Personality and Intention

The confirmation that prominence and responsibility are the key personality dimensions influencing behavioral intentions reveals a critical duality for heritage destinations. The strong positive effect of responsibility underscores a paradigm shift in tourist values, aligning with the growing emphasis on sustainable and ethical consumption (Paskova & Zelenka, 2019). Tourists increasingly reward destinations that demonstrate “care” for their cultural and environmental assets. Conversely, the significant but comparatively less weight of prominence suggests its double-edged nature. While fame attracts visitors, it also carries the risk of over-tourism, which can weaken the experience and negatively impact intentions, as noted by I. Damnjanović (2021). The spatial analysis, showing negative correlations between exceptionality and prominence/responsibility, further hints at this tension: being exceptional might conflict with being widely prominent if not managed responsibly. Therefore, the challenge for heritage managers is not merely to promote prominence but to anchor it in responsibility. Strategies must evolve from generic promotion to managing visitor flows, ensuring authenticity, and transparently communicating conservation efforts, thus transforming potential negatives of fame into a sustainable competitive advantage.

Emotion and Intention

The analysis revealed a powerful impact of negative emotions on intention (β = −.784). This finding strongly supports the negativity theory in tourism, where negative experiences often have a more substantial impact on future behavior than positive ones (Neidhardt et al., 2015; Volo, 2021). The fact that negative emotions, like anger, fear, and sadness, formed a reliable construct in the model indicates that tourists’ dissatisfaction often stems from a combination of these feelings, likely triggered by service failures, overcrowding, or perceived disrespect to the heritage. The weaker direct effect of positive emotion (joy−0.05) suggests that in a heritage context, simply creating joy may not be sufficient to guarantee loyalty; it can be expected as a baseline. The critical task is the active prevention and mitigation of negative experiences. This goes beyond general service quality; it requires a proactive approach to identifying spatial and experiential negative points and addressing them directly. Cities such as George Town should minimize triggers of negative feelings by ensuring reliable, high-quality services, supplying accurate information to shape expectations, and promptly resolving any problems. Identifying context-specific emotional triggers can also guide destination marketing and management; for example, features that repeatedly evoke fear can be targeted for improvement (Schneider, 2010).

Emotion as the Mediator

The mediation analysis provides an insight of this study: negative emotion acts as a full mediator between destination personality and behavioral intention. These findings illuminate the emotional processes underlying tourism behavior. Whereas earlier studies stressed the direct role of destination attributes in shaping intentions (C. F. Chen & Phou, 2013), the present work highlights the influence of a destination’s personality (prominence and responsibility) on a tourist’s decision to return or recommend is almost entirely channeled through the emotional experiences. The non-significant direct path after accounting for the mediator underscores this. This result validates the affective component of the CAC theory in a spatial UGC context. It suggests that a destination’s personality traits are cognitive evaluations that must be felt to influence behavior. For example, a destination’s responsibility must translate into a feeling of safety, respect, or ethical satisfaction for it to positively affect intentions. Conversely, the prominence of a site can quickly lead to negative emotions, if it manifests as overcrowding. Therefore, local marketing cannot just communicate personality traits; it must manage experiences that reliably evoke the positive emotions aligned with those traits and, more importantly, systematically eliminate sources of negative emotion.

Methodological Constructions

This study serves as an example for an integrated UGC-spatial-analytical framework. By combining sentiment analysis, GIS interpolation, and SEM, we moved beyond describing what tourists feel to modeling where and how these feelings interact with perceived destination personality to drive intentions. The use of open-source tools demonstrates a reproducible and effective methodology for DMOs with limited budgets. However, this approach has limitations. The reliance on lexicon-based dictionaries, while robust, may miss nuanced or culturally specific expressions of personality and emotion that advanced natural language processing models could capture. Furthermore, the IDW assumes a smooth gradient of emotion/personality among review points, which may simplify the complex, patchy reality of urban emotional landscapes. Future research should validate this framework in different cultural and destination contexts (e.g., natural reserves) and integrate more advanced techniques like topic modeling to enrich the understanding of why certain emotions arise in specific locations.

Other Findings

After calculating distances across the five destination-personality dimensions (Figure 6), several sites (Church of the Assumption, Wonder Food Museum, Penang Ferry, Little India, Padang Kota Lama, Goddess of Mercy Temple, Chew Jetty, and Fort Cornwallis) show similar profiles. This similarity can be attributed to their shared historical significance, cultural diversity, and distinctive visitor experiences, all of which reflect George Town’s multicultural heritage and shape comparable environmental characteristics. These sites also express common values such as heritage preservation and educational appeal. Nevertheless, each place should cultivate unique traits to differentiate itself, a distinction that is crucial for effective destination marketing and management (Nasution, 2012). Further research should incorporate additional spatial parameters (centrality, connectivity, and others) to deepen understanding of personality patterns.

Figure 6.

Distance map between destinations.

Conclusion

The study on the interplay between destination personality, tourist emotions, and behavioral intentions in the case of George Town, Malaysia, has explained the dynamics of tourism destination management. Utilizing advanced sentiment analysis and emotion scoring models, this research has underscored the significance of destination personality in shaping tourists’ behavioral intentions. The research has identified that destination personality influences less tourists’ emotions. But certain personality traits (prominence and responsibility) are crucial in tourism branding. The analysis of UGC through machine learning models revealed a perspective on how various aspects of destination personality resonate with tourists. Notably, anger/ sadness/ fear can unexpectedly yield negative outcomes, emphasizing the need for emotional management in tourism experiences. Thus, this research contributes to the tourism field by studying the relationship between destination personality, tourist emotions, and behavioral intentions. By addressing its limitations and building on its findings, future work can further the study of these dynamics, ultimately informing more effective and sustainable tourism practices.

Limitations and Future Research

Using UGC may introduce a potential bias toward local tourists, which might not accurately represent the citizens in George Town. Additionally, the sentiment analysis model may not fully capture the complexity of all emotions due to challenges in language processing and tourists’ locations. For future research, there is a clear path toward addressing these limitations. A multi-faceted approach that includes both qualitative and quantitative data sources can provide a more comprehensive understanding of tourists’ experiences, such as longitudinal studies for a better understanding of the evolution of destination personalities and their impact on long-term tourist behavior. Cross-cultural studies would also allow for an advanced understanding of how destination personalities are perceived among different groups or stakeholders. Furthermore, methodological innovation can continue to involve more devices and sensors to monitor, study, and manage tourists’ emotions, such as using eye trackers and participatory GIS.

Practical Implications

The findings of this study offer insights for the sustainable development of tourism destinations. Tourism cities should foster emotional connections with tourists not only from personalities but also from more other perspectives to increase loyalty and word-of-mouth promotion. The negative emotions identified present opportunities for targeted improvements to enhance the overall tourist experience. For example, destinations should move beyond product-focused strategies and co-create experiences that evoke memorable emotional responses, such as immersive storytelling, esthetic appreciation, or social bonding moments. For solving the negativity, some measures are important, such as actively collecting emotional feedback for, post-trip surveys, or informal chats, designing for emotion regulation spaces, quiet zones or reset spaces in crowded attractions, etc. And tourism managers and policymakers can utilize these insights to develop strategies that balance the promotion of destination with the responsible management of tourism’s impact. For example, they can create experiences that highlight local culture and heritage while also implementing measures to control tourist flow and preserve the environment. Engaging local communities more in tourism planning and development will also ensure that the benefits are equitably distributed, fostering a sense of place and sustainability.

Theoretical Implications

Theoretically, the research mainly applied the CAC framework to test variable relationships. Most of the findings were consistent with previous research, confirming the value of CAC in tourism and heritage studies within the Malaysian context. One contribution to the theory is the finding of the relationship between destination personality and tourist emotion. Despite the controversy in previous studies, this study found that the relationship between the two was not strong in the case of heritage cities. This study polarizes emotions, which enriched the original CAC. The study also provided additional observations and explanations, such as the relationship between various emotions and intention. Then, the primary contribution and innovation of this research lie in its methodological approach, introducing an innovative and practical spatial analysis framework. The findings underscore the importance of integrating theoretical constructs with practical applications in tourism research. Future studies are encouraged to build on this work by exploring additional variables and employing advanced spatial technologies to further enrich the understanding of tourist behavior and site management.

Footnotes

ORCID iDs

Poh Ling Tan

Kun Sang

Author Contributions

Weng Shaobin: Conceptualization, Methodology, Writing—original draft. Lin Guiye: Methodology, Writing—review and editing; Xu Lu: Writing—review and editing; Sang Kun.: Conceptualization, Methodology, Writing—review and editing, Supervision. Tan Poh Ling: Literature review, Discussion. All authors have read and agreed to the published version of the manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by Talent Introduction Program of Xihua University (Grant No: Z242013); Sichuan Philosophy; Social Science Foundation Project: Modern Design and Culture Research Center (Grant No: MD23E011); Xiamen University Malaysia Research Fund (Grant no.: XMUMRF/2024-C14/IART/0023).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author.

References

Aaker

J. L.

(1997). Dimensions of brand personality. Journal of Marketing Research, 34(3), 347–356.

Adongo

Choe

J. Y.

Sulemana

S. S.

(2024). Environmentally friendly practices in Macau hotels before and after the COVID-19 pandemic: Hotel executives’ perspectives. Journal of Hospitality and Tourism Insights, 7(4), 2442–2459.

Afthanorhan

Ghazali

P. L.

Rashid

(2021). Discriminant validity: A comparison of CBSEM and consistent PLS using Fornell & Larcker and HTMT approaches. Journal of Physics Conference Series, 1874(1), 012085.

Agarwal

Mittal

Agarwal

Mittal

(2016). Machine learning approach for sentiment analysis. Prominent feature extraction for sentiment analysis. 21–45. Springer Cham.

Agyeiwaah

Adam

Dayour

Badu Baiden

(2021). Perceived impacts of COVID-19 on risk perceptions, emotions, and travel intentions: Evidence from Macau higher educational institutions. Tourism Recreation Research, 46(2), 195–211.

Alzboun

Alhur

Khawaldah

Alshurideh

M. T.

(2023). Assessing gastronomic tourism using machine learning approach: The case of google review. International Journal of Data and Network Science, 7(3), 1131–1142.

Assaker

(2020). Age and gender differences in online travel reviews and user-generated-content (UGC) adoption: Extending the technology acceptance model (TAM) with credibility theory. Journal of Hospitality Marketing & Management, 29(4), 428–449.

Bigné

J. E.

Andreu

Gnoth

(2005). The theme park experience: An analysis of pleasure, arousal and satisfaction. Tourism Management, 26(6), 833–844.

Chen

C. F.

Phou

(2013). A closer look at destination: Image, personality, relationship and loyalty. Tourism Management, 36, 269–278.

10.

Chen

Wang

Zhang

Wei

Lyu

(2024). Does linguistic diversity make destinations more sophisticated? Exploring the effects on destination personality. Journal of Destination Marketing & Management, 31, 100828.

11.

Colneric

Demsar

(2020). Emotion recognition on twitter: Comparative study and training a unison model. IEEE Transactions on Affective Computing, 11(3), 433–446.

12.

Dabbous

Barakat

K. A.

(2020). Bridging the online offline gap: Assessing the impact of brands’ social network content quality on brand awareness and purchase intention. Journal of Retailing and Consumer Services, 53, 101966.

13.

Damnjanović

(2021). Overtourism effects: Positive and negative impacts for sustainable development. In Leal Filho

Azul

A. M.

Brandli

Lange Salvia

Wall

(Eds.), Industry, innovation and infrastructure (pp. 808–818). Springer.

14.

Darda

M. A.

Bhuiyan

M. A. H.

(2022). A structural equation model (SEM) for the socio-economic impacts of ecotourism development in Malaysia. PLoS One, 17(8), e0273294.

15.

Davari

Jang

(2024). Travel-based learning: Unleashing the power of destination curiosity. Journal of Travel & Tourism Marketing, 41(3), 396–417.

16.

Demšar

Curk

Erjavec

Gorup

Č.

Hočevar

Milutinovič

Zupan

(2013). Orange: data mining toolbox in Python. Journal of Machine Learning Research, 14(1), 2349–2353.

17.

Demšar

Zupan

Leban

Curk

(2004). Orange: From experimental machine learning to interactive data mining. In: European conference on principles of data mining and knowledge discovery (pp. 537–539). Springer Berlin Heidelberg.

18.

Dobesova

(2024). Evaluation of Orange data mining software and examples for lecturing machine learning tasks in geoinformatics. Computer Applications in Engineering Education, 32(4), e22735.

19.

Dupain

Novitskaya

(2015). Paris destination image from the point of view of Asian students. In Iscontour 2015 Tourism Research Perspectives: Proceedings of the International Student Conference in Tourism Research (pp. 325–340).

20.

Liechty

Santos

C. A.

Park

(2022). ‘I want to record and share my wonderful journey’: Chinese Millennials’ production and sharing of short-form travel videos on TikTok or Douyin. Current Issues in Tourism, 25(21), 3412–3424.

21.

Ekinci

Hosany

(2006). Destination personality: An application of brand personality to tourism destinations. Journal of Travel Research, 45(2), 127–139.

22.

Freeze

Raschke

R. L.

(2007). An assessment of formative and reflective constructs in IS research. University of St. Gallen.

23.

Ghaderi

Mat Som

A. P.

Henderson

J. C.

(2012). Tourism crises and island destinations: Experiences in Penang, Malaysia. Tourism Management Perspectives, 2, 79–84.

24.

Ghorbanzadeh

Rahehagh

(2021). Emotional brand attachment and brand love: The emotional bridges in the process of transition from satisfaction to loyalty. Rajagiri Management Journal, 15(1), 16–38.

25.

Gnoth

Baloglu

Ekinci

Sirakaya-Turk

(2007). Introduction: Building destination brands. Tourism Analysis, 12(5), 339–343.

26.

Göker

Z. G.

Çelik

A. B.

(2022). Women’s dialogic encounters: Agonistic listening and emotions in multiple-identity conflicts. Third World Quarterly, 43(6), 1251–1269.

27.

Gunn

C. A.

(1972). Vacationscape: Designing Tourist Regions. Taylor & Francis.

28.

Guo

Pesonen

Komppula

(2021). Comparing online travel review platforms as destination image information agents. Information Technology & Tourism, 23, 159–187.

29.

Hair

J. F.

(2006). Multivariate data analysis. Pearson Prentice Hall, 2, 729–764.

30.

Hair

J. F.

Matthews

L. M.

Matthews

R. L.

Sarstedt

(2017). PLS-SEM or CB-SEM: Updated guidelines on which method to use. International Journal of Multivariate Data Analysis, 1(2), 107–123.

31.

M. T.

Nguyen

G. D.

Nguyen

T. H. T.

Thi Nguyen

B. D.

(2023). The use of dietary supplements and vitamin consumption during and after the Covid pandemic in Vietnam: A perspective of user-generated content. Humanities and Social Sciences Communications, 10(1), 1–12.

32.

Hassan

M. A. E.

Zerva

Aulet

(2024). A model for brand personality word embedding: Identifying UNESCO World heritage personality categories. Journal of Travel Research, 63(1), 44–63.

33.

Hooper

Coughlan

Mullen

(2008, June 20). Evaluating model fit: A synthesis of the structural equation modelling literature. In 7th European Conference On Research Methodology For Business And Management Studies, 20th June 2008, London (pp. 195–200).

34.

Hosany

Ekinci

Uysal

(2007). Destination image and destination personality. International Journal of Culture Tourism and Hospitality Research, 1(1), 62–81.

35.

Hosany

Gilbert

(2010). Measuring tourists’ emotional experiences toward hedonic holiday destinations. Journal of Travel Research, 49(4), 513–526.

36.

Hosany

Hunter-Jones

McCabe

(2020). Emotions in tourist experiences: Advancing our conceptual, methodological and empirical understanding. Journal of Destination Marketing & Management, 16, 100444.

37.

Hosany

Prayag

(2013). Patterns of tourists’ emotional responses, satisfaction, and intention to recommend. Journal of Business Research, 66(6), 730–737.

38.

Huang

Scott

Ding

Cheng

(2012). Impression of Liusanjie: Effect of mood on experience and satisfaction. International Journal of Tourism Research, 14(1), 91–102.

39.

Liu

(2004). Mining opinion features in customer reviews. AAAI Conference on Artificial Intelligence, 4(4), 755–760.

40.

Hunt

J. D.

(1975). Image as a factor in tourism development. Journal of Travel Research, 13(3), 1–7.

41.

Janocha

Shpakou

Sobieszczańska

Kałka

Klimatckaia

Woźniak

Pilecka-Radziszewska

Małyszczak

(2018). Emotions and their cognitive and adaptive functions. Progress in Health Sciences, 8, 178–184.

42.

Johnson

J. L.

Arif

Bloom

T. J.

Isaacs

A. N.

Moseley

L. E.

Janke

K. K.

(2023). Preparing pharmacy educators as expedition guides to support professional identity formation in pharmacy education. American Journal of Pharmaceutical Education, 87(1), 8944.

43.

Jovanović

Božić

Dinić

Majstorović

(2017). Serbian adaptation of the Brand Personality Scale (BPS): An application to tourism destinations. Primenjena psihologija, 10(1), 37–61.

44.

Kar

A. K.

(2021). What affects usage satisfaction in mobile payments? Modelling user generated content to develop the “digital service usage satisfaction model. Information Systems Frontiers, 23(5), 1341–1361.

45.

Kastenholz

Marques

C. P.

Carneiro

M. J.

(2020). Place attachment through sensory-rich, emotion-generating place experiences in rural tourism. Journal of Destination Marketing & Management, 17, 100455.

46.

Kim

Ritchie

McCormick

(2012). Development of a scale to measure memorable tourism experiences. Journal of Tourism and Research, 51(1), 12–25.

47.

Kiviniemi

M. T.

Ellis

E. M.

Hall

M. G.

Moss

J. L.

Lillie

S. E.

Brewer

N. T.

Klein

W. M. P.

(2018). Mediation, moderation, and context: Understanding complex relations among cognition, affect, and health behaviour. Psychology and Health, 33(1), 98–116.

48.

Kovacic

Jovanovic

Dinic

(2020). Development and validation of a new measure of travel destination personality. Psihologija, 53(1), 65–85.

49.

Kovačić

Jovanović

Vujičić

M. D.

Morrison

A. M.

Kennell

(2022). What shapes activity preferences? The role of tourist personality, destination personality and destination image: Evidence from Serbia. Sustainability, 14(3), 1803.

50.

Kumar

Nayak

J. K.

(2018). Destination personality: Scale development and validation. Journal of Hospitality & Tourism Research, 42(1), 3–25.

51.

Lam

L. W.

(2012). Impact of competitiveness on salespeople's commitment and performance. Journal of Business Research, 65(9), 1328–1334.

52.

Lee

Cai

De Franco

Lee

(2020). Exploring influential factors affecting guest satisfaction: Big data and business analytics in consumer-generated reviews. Journal of Hospitality and Tourism Technology, 11(1), 137–153.

53.

Zhang

Wang

Chen

(2022). Built environment impacts on rural residents’ daily travel satisfaction. Frontiers in Ecology and Evolution, 10, 931118.

54.

Liu

Bao

Zheng

(2019). Exploring consumers’ purchase intention in social commerce. Asia Pacific Journal of Marketing and Logistics, 31(2), 378–397.

55.

Liu

Shen

(2020). Aspect-based sentiment analysis with gated alternate neural network. Knowledge-Based Systems, 188, 1–14.

56.

Liu

Zheng

(2024). The persuasive power of social media influencers in brand credibility and purchase intention. Humanities and Social Sciences Communications – Nature, 11, 15.

57.

Yen

C. L.

Liu

(2020). Hotel brand personality and brand loyalty: An affective, conative and behavioral perspective. Journal of Hospitality Marketing & Management, 29(5), 550–570.

58.

Masoudi

(2021). Estimation of the spatial climate comfort distribution using tourism climate index (TCI) and inverse distance weighting (IDW) (case study: Fars Province, Iran). Arabian Journal of Geosciences, 14(5), 363.

59.

Mathur

Tewari

Singh

(2022). Modeling the factors affecting online purchase intention: The mediating effect of consumer’s attitude towards user- generated content. Journal of Marketing Communications, 28(7), 725–744.

60.

Mcintosh

A. J.

Siggs

(2005). An exploration of the experiential nature of boutique accommodation. Journal of Travel Research, 44(1), 74–81.

61.

Mehra

(2023). Unexpected surprise: Emotion analysis and aspect based sentiment analysis (ABSA) of user generated comments to study behavioral intentions of tourists. Tourism Management Perspectives, 45, 101063.

62.

Mitra

Mohanty

(2020). Sentiment analysis using machine learning approaches (lexicon based on movie review dataset). Journal of Ubiquitous Computing and Communication Technologies (UCCT), 2(3), 145–152.

63.

Mohaidin

Wei

K. T.

Ali Murshid

(2017). Factors influencing the tourists’ intention to select sustainable tourism destination: A case study of Penang, Malaysia. International Journal of Tourism Cities, 3(4), 442–465.

64.

Moyroud

Portet

(2018). Introduction to QGIS. QGIS and generic tools, 1, 1–17.

65.

Nasution

K. S.

(2012). Exploring shared histories, preserving shared heritage: Penang’s links to a Siamese past. Journal of the Siam Society, 100, 295–322.

66.

Neidhardt

Seyfang

Schuster

Werthner

(2015). A picture-based approach to recommender systems. Information Technology & Tourism, 15(1), 49–69.

67.

Nella

(2022). A review of destination personality literature: focus on articles published in the last decade (2013–2022). In International conference of the international association of cultural and digital tourism (pp. 455–468). Springer International Publishing

68.

Pan

Zhang

Gursoy

(2017). Development and validation of a destination personality scale for mainland Chinese travelers. Tourism Management, 59, 338–348.

69.

Papadimitriou

Apostolopoulou

Kaplanidou

(2015). Destination personality, affective image, and behavioral intentions in domestic urban tourism. Journal of Travel Research, 54(3), 302–315.

70.

Paskova

Zelenka

(2019). How crucial is the social responsibility for tourism sustainability? Social Responsibility Journal, 15(4), 534–552.

71.

Pei

L. T. P.

Nor Azila

M. N.

(2020). The influence of tourist interactions on satisfaction and destination loyalty of tourists at cultural heritage sites in Penang and Melaka. Malaysia, 9(1), 103–119.

72.

Preacher

K. J.

Hayes

A. F.

(2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879–891.

73.

Quintal

Sung

Liu

M. T.

Duong

C. V.

(2024). Hitting it out of the park with park personality: Scale development and validation. Journal of Destination Marketing & Management, 31, 100869.

74.

Rosas-Chavoya

Gallardo-Salazar

J. L.

López-Serrano

P. M.

Alcántara-Concepción

P. C.

León-Miranda

A. K.

(2022). QGIS a constantly growing free and open-source geospatial software contributing to scientific development. Cuadernos de Investigación Geográfica, 48(1), 197–213.

75.

Saadon

J. R.

Yang

Burgert

Mohammad

Gammel

Sepe

Rafailovich

Mikell

C. B.

Polak

Mofakham

(2023). Real-time emotion detection by quantitative facial motion analysis. PLoS One, 18(3), e0282730.

76.

Şahin

İ.

Güzel

F. Ö

. (2020). Do experiential destination attributes create emotional arousal and memory?: A comparative research approach. Journal of Hospitality Marketing & Management, 29, 956–986.

77.

San Martín

Rodríguez del Bosque

I. A

. (2008). Exploring the cognitive–affective nature of destination image and the role of psychological factors in its formation. Tourism Management, 29(2), 263–277.

78.

Scuttari

Volgger

Pechlaner

(2016). Transition management towards sustainable mobility in Alpine destinations: Realities and realpolitik in Italy's South Tyrol region. Journal of Sustainable Tourism, 24(3), 463–483.

79.

Schneider

P. P.

(2010). Exploring the motivation and personality traits of adventure travelers: A hierarchical model approach. Michigan State University.

80.

Soltani

Soltani Nejad

Taheri Azad

Taheri

Gannon

M. J.

(2021). Food consumption experiences: A framework for understanding food tourists’ behavioral intentions. International Journal of Contemporary Hospitality Management, 33(1), 75–100.

81.

Soon

Lim

W. M.

Gaur

S. S.

(2023). The role of emotions in augmented reality. Psychology and Marketing, 40(11), 2387–2412.

82.

Statista. (2014). Emotions of travelers while booking vacations worldwide. Retrieved May 01, 2024 from https://www.statista.com/statistics/324883/emotions-of-travelers-while-booking-vacations-worldwide/

83.

Yang

Swanson

S. R.

Chen

N. C.

(2022). The impact of online reviews on destination trust and travel intention: The moderating role of online review trustworthiness. Journal of Vacation Marketing, 28(4), 406–423.

84.

Sung

Kim

(2010). Effects of brand personality on brand trust and brand affect. Psychology and Marketing, 27(7), 639–661.

85.

Suryaningsih

I. B.

Nugraha

K. S. W.

Anggita

R. M.

(2020). Is cultural background moderating the destination personality and self image congruity relationship of behavioral intention? Jurnal Aplikasi Manajemen, 18(2), 353–362.

86.

Tang

Ren

(2022). Influencing factors and mechanism of tourists’ pro-environmental behavior – Empirical analysis of the CAC-MOA integration model. Frontiers in Psychology, 13, 1060404.

87.

Telisman-Kosuta

(1989). Tourist destination image. In Witt

S. F.

Moutinho

(Eds.), Tourism Marketing and Management Handbook (pp. 557–561). Prentice Hall.

88.

Thomson

MacInnis

D. J.

Whan Park

(2005). The ties that bind: Measuring the strength of consumers’ emotional attachments to brands. Journal of Consumer Psychology, 15(1), 77–91.

89.

Torres-Naranjo

Gálvez

J. C. P.

López-Guzmán

Jara-Alba

(2021). Traveler emotional perceptions in world heritage cities. Tourism Culture & Communication, 21(2), 81–94.

90.

Tran

L. T. T.

(2020). Online reviews and purchase intention: A cosmopolitanism perspective. Tourism Management Perspectives, 35, 100722.

91.

Tran

V. A.

Tran

Q. N.

Doan

H. M.

G. L.

T. D.

Pham

T. H.

(2025). How sensory marketing affects perceived service quality and customer satisfaction: Study in Vietnam public hospitals. Journal of Health Organization and Management, 39, 568–583.

92.

Truong

McColl

(2011). Intrinsic motivations, self-esteem, and luxury goods consumption. Journal of Retailing and Consumer Services, 18(6), 555–561.

93.

UNESCO. (2008). Melaka and George Town, Historic Cities of the Straits of Malacca. Retrieved October 11, 2024 from https://whc.unesco.org/en/list/1223/

94.

Usakli

Baloglu

(2011). Brand personality of tourist destinations: An application of self-congruity theory. Tourism Management, 32, 114–127.

95.

Volo

(2021). The experience of emotion: Directions for tourism design. Annals of Tourism Research, 86, 103097.

96.

Wang

Lin

(2022). Revisit the formation of destination brand personality. Annals of Tourism Research, 95, 103408.

97.

Wang

K. J.

Hsu

C. N.

Sanjaya

(2023). Empowering facial emotion recognition in service industry – AA two-stage convolutional neural network model. Multimedia Tools and Applications, 83, 33161–33184.

98.

Yang

Isa

S. M.

Yao

Xia

Liu

(2022). Cognitive image, affective image, cultural dimensions, and conative image: A new conceptual framework. Frontiers in Psychology, 13, 935814.

99.

Yang

Mohd Isa

Ramayah

(2020). A theoretical framework to explain the impact of destination personality, self-congruity, and tourists’ emotional experience on behavioral intention. Sage Open, 10(4).

100.

Yao

(2013). Assessing tourist experience satisfaction with a heritage destination [Master dissertation]. Purdue University.

101.

C. E.

Zhang

(2020). The embedded feelings in local gastronomy: A sentiment analysis of online reviews. Journal of Hospitality and Tourism Technology, 11(3), 461–478.

102.

Zervas

Proserpio

Byers

J. W.

(2021). A first look at online reputation on Airbnb, where every stay is above average. Marketing Letters, 32, 1–16.

103.

Zhang

Huang

Cao

Chen

(2019). Recognise me from outside to inside: Learning the influence chain of urban destination personalities. Tourism Management, 70, 390–403.

104.

Zhang

Tan

Jiao

Lin

Fan

Xiang

Law

Zheng

(2024). A predictive model based on TripAdvisor textual reviews: Early destination recommendations for travel planning. Sage Open, 14(2).

Does Destination Personality Influence Tourists’ Emotion and Intention? A Study Based on UGC and Spatial Sentiment Analysis

Abstract

Plain Language Summary

Keywords

Introduction

Literature Review

Theoretical Foundation

Hypothesis Development

Methodology

Study Area

Data Collection and Cleaning

Computing Sentiments

Destination Personality

Spatial Analysis

Structural Equation Modeling

Results

Emotion and Personality Distribution

SEM and Mediation Test

Discussions

Variable Relationships

Destination Personality and Emotion

Personality and Intention

Emotion and Intention

Emotion as the Mediator

Methodological Constructions

Other Findings

Conclusion

Limitations and Future Research

Practical Implications

Theoretical Implications

Footnotes

ORCID iDs

Author Contributions

Funding

Declaration of Conflicting Interests

Data Availability Statement

References