Feature-Based Visual Sentiment Analysis of Text Document Streams

Abstract

This article describes automatic methods and interactive visualizations that are tightly coupled with the goal to enable users to detect interesting portions of text document streams. In this scenario the interestingness is derived from the sentiment, temporal density, and context coherence that comments about features for different targets (e.g., persons, institutions, product attributes, topics, etc.) have. Contributions are made at different stages of the visual analytics pipeline, including novel ways to visualize salient temporal accumulations for further exploration. Moreover, based on the visualization, an automatic algorithm aims to detect and preselect interesting time interval patterns for different features in order to guide analysts. The main target group for the suggested methods are business analysts who want to explore time-stamped customer feedback to detect critical issues. Finally, application case studies on two different datasets and scenarios are conducted and an extensive evaluation is provided for the presented intelligent visual interface for feature-based sentiment exploration over time.

Social Media Visual Analytics

“…Compared with the personal emotion analysis, Steed et al [SDB*15] visualize the aggregated emotional dynamics of a large group of people with the high‐dimensional projection. Rohrdantz et al [RHD*12] propose automatic methods and interactive visualizations to extract sentiment from text document streams. The system supports users to analyze sentiment patterns, explore time‐stamped customer feedback and detect critical issues.…”

Event identification for local areas using social media streaming data

“…This method is also used by Rohrdantz et al [17] to identify context-coherence of terms in consumer feedback comments. In contrast to this approach, we build two contingency tables (see Table 1 and Table 2), one for the geographic dimension and one for the time dimension.…”

OpinionBlocks: A Crowd-Powered, Self-improving Interactive Visual Analytic System for Understanding Opinion Text

“…Carenini and Rizoli built a multimedia interface that facilitates the comparison of different reviews [7]. More recently, Huang et al presented RevMiner, an interactive system that summarizes reviews in noun-adjective pairs to be presented in a compact mobile phone interface [13]; and Rohrdantz et al designed a visualization system that supports feature-based sentiment analysis of time-stamped review documents [26]. Similar to these efforts, our work aims at creating summaries of online opinions to help users in their decision-making processes.…”

“…Furthermore, readers often wish to see the concrete evidence behind the extracted aspects and sentiment in a summary [16,26]. OpinionBlocks enables users to "drilldown" through clicking or hovering on the visual elements, allowing them to see snippets associated with blocks, keywords associated with snippets, snippets associated with keywords, or even the full context of the original reviews (Figure 2 Right).…”