Search Results (7)

Social media allows people to express and share a variety of experiences, opinions, beliefs, interpretations, or viewpoints on a single topic. Summarizing a collection of social media posts (microblogs) on one topic may be challenging and can result in an incoherent summary due to multiple perspectives from different users. We introduce a novel approach to microblog summarization, the Multiple-View Summarization Framework (MVSF), designed to efficiently generate multiple summaries from the same social media dataset depending on chosen perspectives and deliver personalized and fine-grained summaries. The MVSF leverages component-of-perspective computing, which can recognize the perspectives expressed in microblogs, such as sentiments, political orientations, or unreliable opinions (fake news). The perspective computing can filter social media data to summarize them according to specific user-selected perspectives. For the summarization methods, our framework implements three extractive summarization methods: Entity-based, Social Signal-based, and Triple-based. We conduct comparative evaluations of MVSF summarizations against state-of-the-art summarization models, including BertSum, SBert, T5, and Bart-Large-CNN, by using a gold-standard BBC news dataset and Rouge scores. Furthermore, we utilize a dataset of 18,047 tweets about COVID-19 vaccines to demonstrate the applications of MVSF. Our contributions include the innovative approach of using user perspectives in summarization methods as a unified framework, capable of generating multiple summaries that reflect different perspectives, in contrast to prior approaches of generating one-size-fits-all summaries for one dataset. The practical implication of MVSF is that it offers users diverse perspectives from social media data. Our prototype web application is also implemented using ChatGPT to show the feasibility of our approach. Full article

Relieving the emotions of the public through government response is an important part of government emergency management. How governments respond in different situations can avoid stimulating negative emotions during emergencies? This paper analyzes the problem from the perspective of configuration; that is, this paper explores the combined effects of multiple factors on emotions. We construct the theoretical framework “Situation-Responder-Content” from situation, responder and response content, and use the government microblogs (n= 1517) from 23 major production accidents in China for the discussion with the use of fuzzy set qualitative comparison analysis (fsQCA). According to the results, the effective response types of different agencies in emergencies are summarized. Local authorities can respond in ways that include “Measures type” and “Measures-Emotion type”. Functional agencies can respond through “Measures type”, “Measures-Emotion type” and “Government feature-Driven” type. This study emphasizes that government response in emergencies is a systematic process. Responsive agencies need to release effective information on the basis of fully considering the situation and other factors. Configuration analysis should also be an important direction in government response research, which is helpful to the practice of government response in emergencies. Full article

Disasters do not follow a predictable timetable. Rapid situational awareness is essential for disaster management. People witnessing a disaster in the same area and beyond often use social media to report, inform, summarize, update, or warn each other. These warnings and recommendations are faster than traditional news and mainstream media. However, due to the massive amount of raw and unfiltered information, the data cannot be managed by humans in time. Automated situational awareness reporting could significantly and sustainably improve disaster management and save lives by quickly filtering, detecting, and summarizing important information. In this work, we aim to provide a novel approach towards automated situational awareness reporting using microblogging data through event detection and summarization. Therefore, we combine an event detection algorithm with different summarization libraries. We test the proposed approach against data from the Russo-Ukrainian war to evaluate its real-time capabilities and determine how many of the events that occurred could be highlighted. The results reveal that the proposed approach can outline significant events. Further research can be carried out to improve short-text summarization and filtering. Full article

With the advent of the Internet era, Chinese users tend to choose to express their opinions on social media platforms represented by Sina Weibo. The changes in people’s emotions toward cities from the microblogging texts can reflect the image of cities presented on mainstream social media, and thus target a good image of cities. In this paper, we collected microblog data containing “Shanghai” from 1 January 2019 to 1 September 2022 by Python technology, and we used three methods: Term Frequency-Inverse Document Frequency keyword statistics, Latent Dirichlet Allocation theme model construction, and sentiment analysis by Zhiwang Sentiment Dictionary. We also explore the impact of the COVID-19 epidemic on Shanghai’s urban image in the context of the “Shanghai Territorial Static Management”, an important public opinion topic during the COVID-19 epidemic. The results of the study show that the “Shanghai-wide static management” of COVID-19 epidemic has significantly reduced the public’s perception of Shanghai and negatively affected the city’s image. By analyzing the data results, we summarize the basic characteristics of Shanghai’s city image and provide strategies for communicating Shanghai’s city image in the post-epidemic era. Full article

With the development of the Web 2.0 era, tourists can freely publish their destination experiences through online travel notes. This enables tourists to become important agents to project tourism destination image (TDI), impacting destination-sustainable development. Previous studies have compared the difference in the images projected by destination management organizations (DMO) and tourists through their published content. However, fewer studies have been done to explore the inter-influences between them on the diachronic process of TDI construction. From the perspective of “circle of representation” this question is researched through a case study of Chiang Mai, Thailand, regarding the market of mainland Chinese tourists. Through interviews and the collection of microblogs from the Thailand National Tourism Bureau and tourists’ travel notes from 2009 to 2021, we found that Chiang Mai has experienced four stages of TDI construction, during which the “Xiao Qingxin” image is evolutionally constructed and formed into the representation circle. The inter-influences between DMO and tourists, as well as the influencing factors in this process, are summarized. Our study supplements a dynamic diachronic analysis of TDI from the constructivism perspective. Relevant management and marketing applications for TDI and destination sustainability in the post-pandemic and Web 2.0 era are also provided. Full article

Text summarization namely, automatically generating a short summary of a given document, is a difficult task in natural language processing. Nowadays, deep learning as a new technique has gradually been deployed for text summarization, but there is still a lack of large-scale high quality datasets for this technique. In this paper, we proposed a novel deep learning method to identify high quality document–summary pairs for building a large-scale pairs dataset. Concretely, a long short-term memory (LSTM)-based model was designed to measure the quality of document–summary pairs. In order to leverage information across all parts of each document, we further proposed an improved LSTM-based model by removing the forget gate in the LSTM unit. Experiments conducted on the training set and the test set built upon Sina Weibo (a Chinese microblog website similar to Twitter) showed that the LSTM-based models significantly outperformed baseline models with regard to the area under receiver operating characteristic curve (AUC) value. Full article

Check-in records are usually available in social services, which offer us the opportunity to capture and analyze users’ spatial and temporal behaviors. Mining such behavior features is essential to social analysis and business intelligence. However, the complexity and incompleteness of check-in records bring challenges to achieve such a task. Different from the previous work on social behavior analysis, in this paper, we present a visual analytics system, Social Check-in Fingerprinting (Sci-Fin), to facilitate the analysis and visualization of social check-in data. We focus on three major components of user check-in data: location, activity, and profile. Visual fingerprints for location, activity, and profile are designed to intuitively represent the high-dimensional attributes. To visually mine and demonstrate the behavior features, we integrate WorldMapper and Voronoi Treemap into our glyph-like designs. Such visual fingerprint designs offer us the opportunity to summarize the interesting features and patterns from different check-in locations, activities and users (groups). We demonstrate the effectiveness and usability of our system by conducting extensive case studies on real check-in data collected from a popular microblogging service. Interesting findings are reported and discussed at last. Full article