Music emotion information is widely used in music information retrieval, music recommendation, music therapy, and so forth. In the field of music emotion recognition (MER), computer scientists extract musical features to identify musical emotions, but this method ignores listeners’ individual differences. Applying machine learning methods, this study formed relations among audio features, individual factors, and music emotions. We used audio features and individual features as inputs to predict the perceived emotion and felt emotion of music, respectively. The results show that real-time individual features (e.g., preference for target music and mechanism indices) can significantly improve the model’s effect, and stable individual features (e.g., sex, music experience, and personality) have no effect. Compared with the recognition models of perceived emotions, the individual features have greater effects on the recognition models of felt emotions.

Response rate has long been a major concern in survey research commonly used in many fields such as marketing, psychology, sociology, and public policy. Based on 244 published survey studies on consumer satisfaction, loyalty, and trust, this study aimed to identify factors that were predictors of response rates. Results showed that response rates were associated with the mode of data collection (face-to-face > mail/telephone > online), type of survey sponsors (government agencies > universities/research institutions > commercial entities), confidentiality (confidential > non-confidential), direct invitation (yes > no), and cultural orientation (individualism > collectivism). A decision tree regression analysis (using classification and regression Tree (C&RT) algorithm on 80% of the studies as the training set and 20% as the test set) revealed that a model with all above-mentioned factors attained a linear correlation coefficient (0.578) between the predicted values and actual values, which was higher than the corresponding coefficient of the traditional linear regression model (0.423). A decision tree analysis (using C5.0 algorithm on 80% of the studies as the training set and 20% as the test set) revealed that a model with all above-mentioned factors attained an overall accuracy of 78.26% in predicting whether a survey had a high (>50%) or low (<50%) response rate. Direct invitation was the most important factor in all three models and had a consistent trend in predicting response rate.

The internet’s convenience and anonymity have facilitated different types of covert fraud, resulting in economic, mental, and social harm to victims. Understanding why people are deceived and implementing appropriate interventions is critical for fraud reduction. Based on the Bayesian brain theory, individuals’ mental states may be a key point in scam compliance and warning compliance. Fraud victims with different mental states may construct various hypotheses and explanations about the fraud they are exposed to, causing different cognition and behavior patterns. Therefore, we first conducted a semi-structured in-depth interview with online fraud victims to investigate the individual and social factors that affect victims’ mental states. Grounded theory analysis showed five core factors influencing scam compliance: psychological traits, empirical factors, motivation, cognitive biases, and emotional imbalance. Based on our findings of psychological processes and deception’s influential factors, we then designed warnings to inform victims of fraud, particularly for those involving novel types of scams. Tested on a real-life setting, our designed warnings effectively enhanced warning compliance, allowing more fraud victims to avoid financial losses.

To make safe road-crossing decisions, it is necessary for pedestrians to accurately estimate the speed and stopping distance of approaching vehicles. Accordingly, the objective of our study was to examine the effects of multiple factors, such as weather conditions, context time (day or night), and illuminance of the roads, on older pedestrians’ (>60 years old) speed perception and stopping distance estimation of approaching vehicles. The participants in this study included 48 older participants who were asked to estimate the speed and stopping distance of approaching vehicles based on 12 s video clips that were selected from natural conditions. The results revealed that actual speeds, weather, context time, and lighting conditions played important roles in the performance of the participants. Compared with young adults, older pedestrians were found to have smaller accurate estimation intervals that varied by multidimensional influencing factors and thus resulted in missing road-crossing opportunities at lower vehicles’ speeds and increasing road-crossing dangers at higher speeds. The older pedestrians’ performance with respect to speed perception and stopping distance estimation is modeled using a regression model with a complex level of tasks. These models can be used by engineers when establishing speed limits and lighting conditions in the areas with senior residents.

As one of the most popular social media platforms in China, Weibo has aggregated huge numbers of texts containing people’s thoughts, feelings, and experiences. Analyzing emotions expressed on Weibo has attracted a great deal of academic attention. Emotion lexicon is a vital foundation of sentiment analysis, but the existing lexicons still have defects such as a limited variety of emotions, poor cross-scenario adaptability, and confusing written and online expressions and words. By combining grounded theory and semi-automatic methods, we built a Weibo-based emotion lexicon for sentiment analysis. We first took a bottom-up approach to derive a theoretical model for emotions expressed on Weibo, and the substantive coding led to eight core emotion categories: joy, expectation, love, anger, anxiety, disgust, sadness, and surprise. Second, we built a new emotion lexicon containing 2,964 words by manually selecting seed words, constructing a word vector model to expand words, and making rules to filter words. Finally, we tested the effectiveness of our lexicon by using a lexicon-based approach to recognize the emotions expressed in Weibo text. The results showed that our lexicon performed better in Weibo emotion recognition than five other Chinese emotion lexicons. This study proposed a method to construct an emotion lexicon that considered both theory and application by combining qualitative research and artificial intelligence methods. Our work also provided a reference for future research in the field of social media sentiment analysis.

Human records can assist us in understanding real descriptions and expected ideals of music. The present work examined how have music emotions been described in millions of Google books. In general, positive adjectives were more regularly used to describe music than negative adjectives, demonstrating a positivity bias in music. The emotional depiction of music has shifted over time, including a decrease in the frequency of emotional adjectives used in English books over the past two centuries, and a sudden surge in the usage of positive adjectives in simplified Chinese books during China’s Cultural Revolution. Negative adjectives were substantially less employed to describe music in simplified Chinese books than in English books, reflecting cultural differences. Finally, a comparison of different corpora showed that emotion-related adjectives were more frequently used to describe music in fictional literature.