Media.Player

Blog Analysis Toolkit

詳細技術說明

None

*Abstract

BackgroundFocus groups, in-depth interviews, content analysis and semiotics are among the many approaches used in qualitative research to gain insight into people's attitudes, behaviors, value systems, concerns, motivations, aspirations, culture or lifestyles. The information that's collected and analyzed - such as blog posts - is unstructured and often text-based. This information can be messy and time consuming to analyze using manual methods. When faced with this type of information - finding themes and extracting meaning can be a daunting task. This requires a solution that is met through Qualitative Data Analysis Systems.TechnologyThe Blog Analysis Toolkit (BAT) is a free, Web-based system for capturing, archiving and sharing blog posts. Blog posts are acquired via RSS feeds, and stored in a database where they can be accessed and shared by other researchers. So far, 414 users have set up BAT accounts since April 2008, and collectively they have archived 104076 posts from 471 blogs. Users can add individual blogs to the repository or do research using samples from existing collections created by other users.Although BAT is a deployed system, it is only a prototype with very limited capabilities. It does not retrieve blog posts from the past; only new posts moving forward in time. Comments left on the blogs, which are potentially very useful for democratic and network theorists, cannot currently be captured, thus it misses the community discussion that makes blogs intrinsically and democratically interesting. For a variety of layout and formatting reasons, BAT cannot currently harvest entire posts on a significant sub-set of the blogosphere, and it also cannot connect important metadata to the text collections. The developers hope to address these limitations as a part of continuing research. Advantages1) Blog aggregation and sampling tools.2) Distributed database construction.3) Data pre-processing to enable simple uploading to Coding Analysis Toolkit (Pitt ID01593).Stage of DevelopmentPrototype software has been developed and successfully tested via end-user group. https://surveyweb2.ucsur.pitt.edu/qblog/page_login.php

*Principal Investigation

Name: Stuart Shulman

Department: SIS-Library & Information Science

國家/地區

美國