Daqing He

Open user profiles for adaptive news systems: help or harm?

Over the last five years, a range of projects have focused on progressively more elaborated techniques for adaptive news delivery. However, the adaptation process in these systems has become more complicated and thus less transparent to the users. In this paper, we concentrate on the application of open user models in adding transparency and controllability to adaptive news systems. We present a personalized news system, YourNews, which allows users to view and edit their interest profiles, and report a user study on the system. Our results confirm that users prefer transparency and control in their systems, and generate more trust to such systems. However, similar to previous studies, our study demonstrate that this ability to edit user profiles may also harm the system.s performance and has to be used with caution.

User-assisted query translation for interactive cross-language information retrieval

Interactive Cross-Language Information Retrieval (CLIR), a process in which searcher and system collaborate to find documents that satisfy an information need regardless of the language in which those documents are written, calls for designs in which synergies between searcher and system can be leveraged so that the strengths of one can cover weaknesses of the other. This paper describes an approach that employs user-assisted query translation to help searchers better understand the systems operation. Supporting interaction and interface designs are introduced, and results from three user studies are presented. The results indicate that experienced searchers presented with this new system evolve new search strategies that make effective use of the new capabilities, that they achieve retrieval effectiveness comparable to results obtained using fully automatic techniques, and that reported satisfaction with support for cross-language searching increased. The paper concludes with a description of a freely available interactive CLIR system that incorporates lessons learned from this research.

Searching, browsing, and clicking in a search session: changes in user behavior by task and over time

There are many existing studies of user behavior in simple tasks (e.g., navigational and informational search) within a short duration of 1--2 queries. However, we know relatively little about user behavior, especially browsing and clicking behavior, for longer search session solving complex search tasks. In this paper, we characterize and compare user behavior in relatively long search sessions (10 minutes; about 5 queries) for search tasks of four different types. The tasks differ in two dimensions: (1) the user is locating facts or is pursuing intellectual understanding of a topic; (2) the user has a specific task goal or has an ill-defined and undeveloped goal. We analyze how search behavior as well as browsing and clicking patterns change during a search session in these different tasks. Our results indicate that user behavior in the four types of tasks differ in various aspects, including search activeness, browsing style, clicking strategy, and query reformulation. As a search session progresses, we note that users shift their interests to focus less on the top results but more on results ranked at lower positions in browsing. We also found that results eventually become less and less attractive for the users. The reasons vary and include downgraded search performance of query, decreased novelty of search results, and decaying persistence of users in browsing. Our study highlights the lack of long session support in existing search engines and suggests different strategies of supporting longer sessions according to different task types.

Analysing Web Search Logs to Determine Session Boundaries for User-Oriented Learning

Incremental learning approaches based on user search activities provide a means of building adaptive information retrieval systems. To develop more effective user-oriented learning techniques for the Web, we need to be able to identify a meaningful session unit from which we can learn. Without this, we run a high risk of grouping together activities that are unrelated or perhaps not from the same user. We are interested in detecting boundaries of sequences between related activities (sessions) that would group the activities for a learning purpose. Session boundaries, in Reuters transaction logs, were detected automatically. The generated boundaries were compared with human judgements. The comparison confirmed that a meaningful session threshold for establishing these session boundaries was confined to a 11-15 minute range.

The state of iSchools: an analysis of academic research and graduate education

The emergence of the iSchool movement and the establishment of iSchools have helped to reshape the landscape of the library and information science (LIS) discipline. In this article, based on a set of research questions focusing around the research and education efforts of about 25 iSchools, we performed a study using both quantitative and qualitative methods on publically available data obtained from the web. Our results show that iSchools share the same vision and mission of working on relationships between information, people and technology, and have established themselves as the appropriate institutions for researchers from diverse subject areas to study this interdisciplinary integration. Overall, we are seeing an emerging iSchool identity and a defining iField, but there are still many important developments to make.

Making MIRACLEs: Interactive translingual search for Cebuano and Hindi

Searching is inherently a user-centered process; people pose the questions for which machines seek answers, and ultimately people judge the degree to which retrieved documents meet their needs. Rapid development of interactive systems that use queries expressed in one language to search documents written in another poses five key challenges: (1) interaction design, (2) query formulation, (3) cross-language search, (4) construction of translated summaries, and (5) machine translation. This article describes the design of MIRACLE, an easily extensible system based on English queries that has previously been used to search French, German, and Spanish documents, and explains how the capabilities of MIRACLE were rapidly extended to accommodate Cebuano and Hindi. Evaluation results for the cross-language search component are presented for both languages, along with results from a brief full-system interactive experiment with Hindi. The article concludes with some observations on directions for further research on interactive cross-language information retrieval.

Comparing user-assisted and automatic query translation

For the 2002 Cross-Language Evaluation Forum Interactive Track, the University of Maryland team focused on query formulation and reformulation. Twelve people performed a total of forty eight searches in the German document collection using English queries. Half of the searches were with user-assisted query translation, and half with fully automatic query translation. For the user-assisted query translation condition, participants were provided two types of cues about the meaning of each translation: a list of other terms with the same translation (potential synonyms), and a sentence in which the word was used in a translation-appropriate context. Four searchers performed the official iCLEF task, the other eight searched a smaller collection. Searchers performing the official task were able to make more accurate relevance judgments with user-assisted query translation for three of the four topics. We observed that the number of query iterations seems to vary systematically with topic, system, and collection, and we are analyzing query content and ranked retrieval measures to obtain further insight into these variations in search behavior.

Toward a Robust data fusion for document retrieval

This paper describes an investigation of signal boosting techniques for post-search data fusion, where the quality of the retrieval results involved in fusion may be low or diverse. The effectiveness of data fusion techniques in such situation depends on the ability of the fusion techniques to be able to boost the signals from relevant documents and reduce the effect of noise that often comes from low quality retrieval results. Our studies on Malach spoken document collection and HARD collection have demonstrated that CombMNZ, the most widely used data fusion method, does not have such ability. We, therefore, developed two versions of signal boosting mechanisms on top of CombMNZ, which result in two new fusion methods called WCombMNZ and WCombMWW. To examine the effectiveness of the two new methods, we conducted experiments on Malach and HARD document collections. Our results show that the new methods can significantly outperform CombMNZ in combining retrieval results that are low and diverse. When the tasks are to combine retrieval results that are in similar quality, which have been the scenarios that CombMNZ are applied often, the two new methods still can obtain often better, sometimes significantly, fusion results.

Modeling search processes using hidden states in collaborative exploratory web search

Investigations of search processes that involve complex interactions, such as collaborative search processes, are important research topics. Previous approaches of directly applying individual search process models into collaborative settings have proven to be problematic. In this paper, we proposed an innovative approach to model collaborative search processes using Hidden Markov Model (HMM), which is an automatic technique for analyzing temporal sequential data. Obtained through a user study, the data used in this paper consist of two different tasks in both collaborative exploratory Web search and individual exploratory Web search conditions. Our results showed that the identified hidden patterns of search process through HMM are compatible with previous well-known models. In addition, HMM generates detailed information on the transitions of hidden patterns in search processes, which demonstrated to be useful for analyzing task differences, and for determining the correlation of search process with search performance. The findings can be used for evaluating collaborative search systems as well as providing guidance for the system design.

Eating disorder questions in Yahoo! Answers: Information, conversation, or reflection?

This study investigated a particular form of social Q&A – Yahoo! Answers – and the nature of the questions posed by teens on the topic of eating disorders. The goals of this study were to identify the information needs of questioners in Yahoo! Answers vis a vis the topic of eating disorders, to create a taxonomy of question types in Yahoo! Answers on the topic of eating disorders, and finally, to contribute to broader models of question-asking in social Q&A. We achieved these goals through a content analysis of 2230 questions posed in Yahoo! Answers between December 2005 and April 2011. This resulted in a classification scheme with five overarching themes - Seeking Information, Seeking Emotional Support, Seeking Communication, Seeking Self-Expression, and Seeking Help to Complete a Task, and the sub-categories of factual, diagnosis, treatment or intervention, validation, seeking comfort, conversation starters, deep talk, confession, reflection, homework help, and manuscript ideas. Through the investigation of the socio-emotional aspects of social Q&A, this study enriches our understanding of the affective dimension of health information behavior.