This study aims to construct a sentiment series generation method for danmu comments based on deep learning, and explore the features of sentiment series after clustering.
Abstract
Purpose
This study aims to construct a sentiment series generation method for danmu comments based on deep learning, and explore the features of sentiment series after clustering.
Design/methodology/approach
This study consisted of two main parts: danmu comment sentiment series generation and clustering. In the first part, the authors proposed a sentiment classification model based on BERT fine-tuning to quantify danmu comment sentiment polarity. To smooth the sentiment series, they used methods, such as comprehensive weights. In the second part, the shaped-based distance (SBD)-K-shape method was used to cluster the actual collected data.
Findings
The filtered sentiment series or curves of the microfilms on the Bilibili website could be divided into four major categories. There is an apparently stable time interval for the first three types of sentiment curves, while the fourth type of sentiment curve shows a clear trend of fluctuation in general. In addition, it was found that “disputed points” or “highlights” are likely to appear at the beginning and the climax of films, resulting in significant changes in the sentiment curves. The clustering results show a significant difference in user participation, with the second type prevailing over others.
Originality/value
Their sentiment classification model based on BERT fine-tuning outperformed the traditional sentiment lexicon method, which provides a reference for using deep learning as well as transfer learning for danmu comment sentiment analysis. The BERT fine-tuning–SBD-K-shape algorithm can weaken the effect of non-regular noise and temporal phase shift of danmu text.
Details
Keywords
Zhou Zhang, Xiaoping Li, Jie Xiong, Jie Yan, Lu Xu and Ruoxi Wang
In the ongoing Industry 4.0 era, the internet of things (IoT) has become a global race in the current information technology climate. However, little is understood about the…
Abstract
Purpose
In the ongoing Industry 4.0 era, the internet of things (IoT) has become a global race in the current information technology climate. However, little is understood about the pattern of the global competitive arena or its players’ set up strategy. This paper aims to attempt to compare the cross-country development of the IoT industry. In particular, from the lens of industrial policies, this paper highlights how China, as a latecomer, gains momentum to emerge victorious as a leader in this global race.
Design/methodology/approach
Based on five dimensions, namely, foundation, trajectory, characteristic, application and social impacts, this paper presents the evolution of the IoT industry in the USA, European Union, Japan, South Korea and China. From the lens of windows of opportunities, this paper analyzes how China seized the opportunity with the emerging technology, thereby, enabling it to create a competitive advantage.
Findings
This paper finds that China’s IoT industry takes a distinct trajectory, where scientific institutions, enterprises and governmental policies collaborate in unison, during which the first phase was when scientific research institutions introduced the conceptual new technology from developed countries. This technological foresight allowed for the identification and realization of critical technologies, strategic fields and technological trends. The second phase was the continuous dissatisfaction of capabilities of critical technologies, which creates disruptions that significantly altered the environment of technological competition.
Originality/value
This paper provides a comprehensive and comparative review of IoT industries in a global context, with the critical and influential role of the windows of opportunities on those enterprises lagging behind the technological wave.
Details
Keywords
Baoku Li, Ruoxi Yao and Yafeng Nan
Benefiting from the development and innovation of artificial intelligence and affective computing technology, social chatbots that integrate cognitive analysis and affective…
Abstract
Purpose
Benefiting from the development and innovation of artificial intelligence and affective computing technology, social chatbots that integrate cognitive analysis and affective social services have flooded into the consumer market. For cognition and emotion-oriented tasks, social chatbots do not always receive positive consumer responses. In addition, consumers have a contradictory attitude toward the anthropomorphism of chatbots. Therefore, from the perspective of mind perception and the two dimensions of social judgment, this research explores the mechanism of consumer responses to anthropomorphic interaction styles when social chatbots complete different service tasks.
Design/methodology/approach
This paper utilizes three behavior experimental designs and survey methods to collect data and the ANOVA, t-test and bootstrap analysis methods to verify the assumed hypotheses.
Findings
The results indicate that when the service task type of a social chatbot is cognition-oriented, compared to a warm anthropomorphic interaction style, a competent anthropomorphic interaction style can improve consumer responses more effectively. During this process, agent-mind perception plays a mediating role. When the service task type of a social chatbot is emotion-oriented, compared with a competent anthropomorphic conversation style, a warm anthropomorphic conversation style can improve consumer responses. Experience-mind perception mediates this influencing relationship.
Originality/value
The research results theoretically enrich the relevant research on the anthropomorphism of social chatbots and expand the application of the theory of mind perception in the fields of artificial intelligence and interactive marketing. Our findings provide theoretical guidance for the anthropomorphic development and design of social chatbots and the practical management of service task scenarios.
Details
Keywords
Baoku Li, Yafeng Nan and Ruoxi Yao
The purpose of this paper is to explore the effect of cuteness and cool on the perceived quality of digital products, the mediating effect of brand perception (warmth and…
Abstract
Purpose
The purpose of this paper is to explore the effect of cuteness and cool on the perceived quality of digital products, the mediating effect of brand perception (warmth and competence) and the moderating effect of the individual perception level.
Design/methodology/approach
This paper utilizes experimental design and survey methods to collect data and the ANOVA, independent sample t-test and bootstrap analysis methods to verify the assumed hypotheses.
Findings
Studies 1 and 2 demonstrate that cuteness (vs cool) is more likely to promote the perception of brand warmth (vs competence), and the brand perception plays a mediating role between cuteness (cool) and the perceived quality. Study 3 replicates the findings of Study 2 and indicates that people with high-cuteness (vs low-cuteness) perception are the same to perceive the brand warmth to promote the perceived quality of digital products, but people with high-cool (vs low-cool) perception are more likely to perceive the brand competence to promote the perceived quality of digital products.
Practical implications
Based on the conclusions in this paper, marketers could emphasize the cool information of digital products in advertisements to promote the perceived quality to promote younger consumers' willingness to pay (WTP). Furthermore, firms could shape warm brand images by the perception of cuteness because cuteness is positively associated with the warmth of brand perception (e.g. the logo of Three Squirrels, a Chinese nut business brand that consists of three cute squirrels).
Originality/value
From a theoretical standpoint, this paper contributes to the brand perception when consumers accept product information with the characteristics of cuteness or cool. Second, a model of perceived quality of digital products is built based on the stereotype content theory. Third, this paper considers individual perception levels on cuteness and cool as the boundaries to conduct further conceptual model.
Details
Keywords
Vahid Nikpey Pesyan, Yousef Mohammadzadeh, Ali Rezazadeh and Habib Ansari Samani
The study aims to examine the impact of cultural dependency stemming from exchange rate fluctuations (specifically the US dollar) on herding behavior in the housing market across…
Abstract
Purpose
The study aims to examine the impact of cultural dependency stemming from exchange rate fluctuations (specifically the US dollar) on herding behavior in the housing market across 31 provinces of Iran from Q2 2011 to Q1 2022, using a spatial econometrics approach. After confirming the presence of spatial effects, the Dynamic Spatial Durbin Panel Model with Generalized Common Effects (SDM-DPD(GCE)) was selected from various spatial models for these provinces.
Design/methodology/approach
The study examines the impact of cultural dependency stemming from exchange rate fluctuations (specifically the US dollar) on herding behavior in the housing market across 31 provinces of Iran from Q2 2011 to Q1 2022, using a spatial econometrics approach. After confirming the presence of spatial effects, the Dynamic Spatial Durbin Panel Model with Generalized Common Effects (SDM-DPD(GCE)) was selected from various spatial models for these provinces.
Findings
The model estimation results indicate that fluctuations in the free market exchange rate of the dollar significantly and positively impact the housing market in both target and neighboring regions, fostering herding behavior characterized by cultural dependency within the specified timeframe. Additionally, the study found that variables such as the inflation rate, population density index and the logarithm of stock market trading volume have significant and positive impacts on the housing market. Conversely, the variable representing the logarithm of the distance from the provincial capital, Tehran, significantly and negatively impacts the housing market across Iranian provinces.
Originality/value
Given that housing is a fundamental need for households, the dramatic price increases in this sector (for instance, a more than 42-fold increase from 2011–2021) have significantly impacted the welfare of Iranian families. Currently, considering the average housing price in Tehran is around 50 million Tomans, and the average income of worker and employee groups is 8 million Tomans (as of 2021), the time required to purchase a 100-square-meter house, even with a 30% savings rate and stable housing prices, is approximately 180 years. Moreover, the share of housing and rent expenses in household budgets now constitutes about 70%. The speculative behavior in this market is so acute that, despite 25 million of Iran’s 87 million population being homeless or renting, over 2.5 million vacant homes (12% of the total housing stock) are not used. Therefore, various financial behaviors and decisions affect Iran’s housing market. Herd behavior is triggered by the signal of national currency devaluation (with currency exchange rates increasing more than 26-fold between 2011 and 2021) and transactions at higher prices in certain areas (particularly in northern Tehran) (Statistical Center of Iran, 2023). Given the origins of housing price surges, a price increase in one area quickly spreads to other regions, resulting in herd behavior in those areas (spillover effect). Consequently, housing market spikes in Iran tend to follow episodes of currency devaluation. Therefore, considering the presented discussions, one might question whether factors other than economic ones (such as herd behavior influenced by dependence culture) play a role in the rising housing prices. Or, if behavioral factors were indeed contributing to the increase in housing prices, what could be the cause of this herd movement? Has the exchange rate, particularly fluctuations in the free market dollar rate, triggered herd behavior in the housing market across Iran’s provinces? Or has the proximity and neighborhood effect been influential in the increase or decrease in housing prices in the market?
Details
Keywords
Zongshui Wang, Wei Liu, Zhuo Sun and Hong Zhao
Building on social media and destination brand-related literature, this study aims to explore World Heritage Sites’ (WHSs) brand diffusion and formation process from long-term and…
Abstract
Purpose
Building on social media and destination brand-related literature, this study aims to explore World Heritage Sites’ (WHSs) brand diffusion and formation process from long-term and short-term perspectives, which includes brand diffusion, user-generated content (UGC), opinion leaders and brand events’ impact.
Design/methodology/approach
This study uses a mixed-method including text mining, keyword analysis and social network analysis to explore the brand formation process of four popular WHSs in Beijing, namely, the Palace Museum, Great Wall, Summer Palace and Temple of Heaven and more than 10,000,000 users’ data on Sina Weibo has been implemented to uncover the underlying social media branding mechanism.
Findings
The results show that the number of postings keeps in a stable range in most months, but, in general, there are no common rules for changing trends among the four WHSs; long-term high-frequency keywords related to history and culture account for a higher percentage; different kinds of accounts have varying impacts on information diffusion, in which media accounts lead to a bigger influence. However, more followers do not necessarily mean more interactions and most of the interaction ratio is much lower than 0.01000; brand events facilitate brand dissemination and have an impact on the creation of UGC.
Practical implications
This study is valuable for destination marketers to deeper understand brand diffusion and formation and provides valuable insights for developing effective destination marketing strategies.
Originality/value
Unlike previous studies that only concern a few parts of destination brand formation via social media (e.g. brand diffusion, brand events or opinion leaders’ impact), this study takes a more comprehensive perspective by systematically analyzing the brand formation process of WHSs on social media. By considering both long-term diffusion and short-term representative events, this study provides a more holistic understanding of the branding mechanism.