visualizing the affective structure of a text document

abstractThis paper introduces an approach for graphically visualiz-ing the affective structure of a text document. A document is first affectively analyzed using a unique textual affect sensing engine, which leverages commonsense knowledge to classify text more reliably and comprehensively than can be achieved with keyword spotting methods alone. Using this engine, sentences are annotated using six basic Ekman emotions. Colors used to represent each of these emotions are sequenced into a color bar, which represents the pro-gression of affect through a text document. Smoothing techniques allow the user to vary the granularity of the af-fective structure being displayed on the color bar. The bar is hyperlinked in a way such that it can be used to easily navigate the document. A user evaluation demonstrates that the proposed method for visualizing and navigating a document’s affective structure facilitates a user’s within-document information foraging activity.

INTRODUCTIONVisualizing a document’s structure is proven to be useful in aiding information retrieval. The classic example is the table of contents, often found at the beginning of a book or lengthy document. In a table of contents, entries are the-matic units such as chapters, sections, and subsections. Glancing through the table, a person can quickly glean the overall thematic structure of the document, allowing him/her to quickly access a section of interest.
Thematic structure, however, is only one of the ways in which people can naturally conceptualize a document. Another way by which people mentally index a document is in terms of its affective structure. In talking about sto-ries, people often make references such as “the sad part” or “the climax.” These are references to a story’s affective structure. This paper introduces an approach for visualiz-ing a document’s affective structure represented in a color-based navigation bar.
First, we describe how the affective structure of text is pro-duced through analysis. Second, we explain the choice of color as the mode of representation. Third, we describe how the color bar is used to navigate a document. Fourth, we discuss a user evaluation. We conclude with a sum-mary of contribution and directions for future work.

기존의 toc(table of content)를 이용한 문서의 인덱싱에서 벗어나 단어를 이용한 감성적 텍스트 분석을 위한 연구. 문장 안에서 sad, angry, fearful, disgusted, surprised, neutral 등의 감정상태를 분류하여 문서 안에 존재하는 감정(감성) 구조를 시각화한다.