Tracking the Evolution of Social Emotions: A Time-Aware Topic Modeling Perspective

Abstract

Many of today's online news websites have enabled users to specify different types of emotions (e.g., angry and shocked) they have after reading news. Compared with traditional user feedbacks such as comments and ratings, these specific emotion annotations are more accurate for expressing users' personal emotions. In this paper, we propose to exploit these users' emotion annotations for online news in order to track the evolution of emotions, which plays an important role in various online services. A critical challenge is how to model emotions with respect to time spans. To this end, we propose a time-aware topic modeling perspective for solving this problem. Specifically, we first develop a model named emotion-Topic over Time (eToT), in which we represent the topics of news as a Beta distribution over time and a multinomial distribution over emotions. While eToT can uncover the latent relationship among news, emotion and time directly, it cannot capture the dynamics of topics. Therefore, we further develop another model named emotionbased Dynamic Topic Model (eDTM), where we explore the state space model for tracking the dynamics of topics. In addition, we demonstrate that both eToT and eDTM could enable several potential applications, such as emotion prediction, emotion-based news recommendations and emotion anomaly detections. Finally, we validate the proposed models with extensive experiments with a real-world data set.

Emotion evolutions of sub-topics about popular events on microblogs

“…Wang et al (2011) investigated hashtag-level sentiment classification on Twitter. Zhu et al (2014) proposed a time-aware topic modelling approach that models emotions with respect to news.…”

“…In this paper, the authors extract sub-topics about a popular event from microblog corpora and identify how emotions about sub-topics evolve over time to obtain more comprehensive information about users' attitudes. Unlike the work of Zhu et al (2014), which focuses on the news corpus and uses manually tagged emotion labels, this paper extracts event sub-topics from social media platforms and EL 35,4 identifies emotion evolutions of sub-topics automatically, using subjectivity classification and emotion classification algorithms.…”

Approval network: a novel approach for sentiment analysis in social networks

“…During the last decade, sentiment analysis has been mainly focused on well-formed text at different granularity such as document level [1,62,69], sentence level [41,68] and aspect level [16,20,72]. In the recent years, the informal texts on social networks become one of the major forms of online communications, enabling a quasi real-time diffusion of contents provided by people and organisations.…”

Advances in Crowd Analysis for Urban Applications Through Urban Event Detection

“…These data are stored and arranged in the database based on keywords (e.g., with Hashtags). These data have been applied to identify crowd events [76], [77], [78], [79], [91], transportation planning and management [21], and crowd sentiment analysis [21], [12], [23].…”

“…Devices enabled with pervasive computing technologies, such as, smart cards [1], [2], wearable devices [3], Radio-Frequency (RF) communication devices (RFCD) (e.g., mobile phones [4], [5], [6], Bluetooth [7], [8], [9], [10], Wireless Fidelity (WiFi) [11], [12], [13], RadioFrequency ID (RFID), and Global Positioning System (GPS) [14], [15], [16]), Optical-Wireless communication (OWC) (e.g., infrared or IR devices) [17], video surveillance [18], [19], [20], in conjunction with social media [21], [22], [23] and different event websites are part and parcel of our daily lives. Recent studies report that the above mentioned ubiquitous technologies can also be employed as sensors to collect data on human activities from the urban space and are uploaded to distributed / centralized databases.…”