β Scribed by Nikhil Sharma; Tapan Khopkar; Jahna Otterbacher
No coin nor oath required. For personal study only.
Web 2.0 has allowed online shoppers to become not just information seekers but also information providers. Many eβcommerce venues allow users to share their experiences in the form of textual product reviews. While textual product reviews represent a wealth of information for candidate buyers, finding pertinent information becomes difficult, as the number of reviews for a particular product becomes large, or if the buyer is interested in particular features of an item. We present ChatterCrop, a tool that uses text summarization. ChatterCrop condenses the information from a large set of reviews into a few sentences, and allows users to customize these summaries for featureβfocused search. In a user study, subjects used ChatterCrop and a sortable list of reviews, to answer questions about two camcorder models. When looking for information about particular features, ChatterCrop outperformed the list, in terms user confidence and perceived ease in finding pertinent information. ChatterCrop's summaries also provided users with starting points to guide further research, and were particularly helpful in the early stages of sensemaking.
π SIMILAR VOLUMES
Operational inefficiencies, sub-optimal patient care, financial losses and missing or ineffective sanctions for non-compliance are all potential outcomes that can stem from passive policy and procedure management. Automation can help solve these problems and provide strategic, operational and financ
Human-centred design processes for interactive systems are de"ned in ISO 13407 and the associated ISO TR 18529. The publication of these standards represents a maturing of the discipline of user-centred design. The systems development community see that (at last) Human Factors has processes which ca
An analysis of data from the first decade of the publication of Online and Online Review revealed that while both cover the literature of online searching, they are very different. Editorial policies, subtitles, and editorial board memberships all indicate differing perspectives. Further, objective
One hundred twenty consumers (classified as either higher or lower knowledge) read an advertisement for a camera described by either attributes or consequences, or both attributes and their related consequences. Thought listings coded for inferences at varying levels of abstraction corresponding to