Ranking related entities components and analyses.

On the Evaluation of Entity Profiles.

The University of Amsterdam at WePS3.

Category%2DBased Query Modeling for Entity Search.
Abstract. Users often search for entities instead of documents and in this setting are willing to provide extra input, in addition to a query, such as category information and example entities. We propose a general probabilistic framework for entity search to evaluate and provide insight in the many ways of using these types of input for query modeling. We focus on the use of category information and show the advantage of a category-based representation over a term-based representation, and also demonstrate the effectiveness of category-based expansion using example entities. Our best performing model shows very competitive performance on the INEX-XER entity ranking and list completion tasks. 1
A two%2Dstage model for blog feed search.

Contextual factors for finding similar experts.

Using Contextual Information to Improve Search in Email Archives.
Abstract. In this paper we address the task of finding topically relevant email messages in public discussion lists. We make two important observations. First, email messages are not isolated, but are part of a larger online environment. This context, existing on different levels, can be incorporated into the retrieval model. We explore the use of thread, mailing list, and community content levels, by expanding our original query with term from these sources. We find that query models based on contextual information improve retrieval effectiveness. Second, email is a relatively informal genre, and therefore offers scope for incorporating techniques previously shown useful in searching user-generated content. Indeed, our experiments show that using query-independent features (email length, thread size, and text quality), implemented as priors, results in further improvements. 1
Combining Term%2DBased and Category%2DBased Representations for Entity Search.

A language modeling framework for expert finding.
 
Non%2Dlocal evidence for expert finding.
The task addressed in this paper, finding experts in an enterprise setting, has gained in importance and interest over the past few years. Commonly, this task is approached as an association finding exercise between people and topics. Existing techniques use either documents (as a whole) or proximity-based techniques to represent candidate experts. Proximity-based techniques have shown clear precision-enhancing benefits. We complement both document and proximity-based approaches to expert finding by importing global evidence of expertise, i.e., evidence obtained using information that is not available in the immediate proximity of a candidate expert’s name occurrence or even on the same page on which the name occurs. Examples include candidate priors, query models, as well as other documents a candidate expert is associated with. Using the CERC data set created for the TREC 2007 Enterprise track we identify examples of non-local evidence of expertise. We then propose modified expert retrieval models that are capable of incorporating both local (either document or snippet-based) evidence and non-local evidence of expertise. Results show that our refined models significantly outperform existing state-of-the-art approaches.
Associating People and Documents.
Abstract. Since the introduction of the Enterprise Track at TREC in 2005, the task of finding experts has generated a lot of interest within the research community. Numerous models have been proposed that rank candidates by their level of expertise with respect to some topic. Common to all approaches is a component that estimates the strength of the association between a document and a person. Forming such associations, then, is a key ingredient in expertise search models. In this paper we introduce and compare a number of methods for building documentpeople associations. Moreover, we make underlying assumptions explicit, and examine two in detail: (i) independence of candidates, and (ii) frequency is an indication of strength. We show that our refined ways of estimating the strength of associations between people and documents leads to significant improvements over the state-of-the-art in the end-toend expert finding task. 1
Finding Key Bloggers%2C One Post At A Time.
Abstract. User generated content in general, and blogs in particular, form an interesting and relatively little explored domain for mining knowledge. We address the task of blog distillation: to find blogs that are principally devoted to a given topic, as opposed to blogs that merely happen to discuss the topic in passing. Working in the setting of statistical language modeling, we model the task by aggregating a blogger’s blog posts to collect evidence of relevance to the topic and persistence of interest in the topic. This approach achieves state-ofthe-art performance. On top of this baseline, we extend our model by incorporating a number of blog-specific features, concerning document structure, social structure, and temporal structure. These blogspecific features yield further improvements. 1
Associating People and Documents.
Abstract. Since the introduction of the Enterprise Track at TREC in 2005, the task of finding experts has generated a lot of interest within the research community. Numerous models have been proposed that rank candidates by their level of expertise with respect to some topic. Common to all approaches is a component that estimates the strength of the association between a document and a person. Forming such associations, then, is a key ingredient in expertise search models. In this paper we introduce and compare a number of methods for building documentpeople associations. Moreover, we make underlying assumptions explicit, and examine two in detail: (i) independence of candidates, and (ii) frequency is an indication of strength. We show that our refined ways of estimating the strength of associations between people and documents leads to significant improvements over the state-of-the-art in the end-toend expert finding task. 1
A Generative Language Modeling Approach for Ranking Entities.

A Generative Language Modeling Approach for Ranking Entities.

A few examples go a long way constructing query models from elaborate query formulations.
We address a specific enterprise document search scenario, where the information need is expressed in an elaborate manner. In our scenario, information needs are expressed using a short query (of a few keywords) together with examples of key reference pages. Given this setup, we investigate how the examples can be utilized to improve the end-to-end performance on the document retrieval task. Our approach is based on a language modeling framework, where the query model is modified to resemble the example pages. We compare several methods for sampling expansion terms from the example pages to support query-dependent and query-independent query expansion; the latter is motivated by the wish to increase “aspect recall, ” and attempts to uncover aspects of the information need not captured by the query. For evaluation purposes we use the CSIRO data set created for the TREC 2007 Enterprise track. The best performance is achieved by query models based on query-independent sampling of expansion terms from the example documents.
Bloggers as experts feed distillation using expert retrieval models.
We address the task of (blog) feed distillation: to find blogs that are principally devoted to a given topic. The task may be viewed as an association finding task, between topics and bloggers. Under this view, it resembles the expert finding task, for which a range of models have been proposed. We adopt two language modelingbased approaches to expert finding, and determine their effectiveness as feed distillation strategies. The two models capture the idea that a human will often search for key blogs by spotting highly relevant posts (the Posting model) or by taking global aspects of the blog into account (the Blogger model). Results show the Blogger model outperforms the Posting model and delivers state-of-the art performance, out-of-the-box.
Parsimonious relevance models.
We describe a method for applying parsimonious language models to re-estimate the term probabilities assigned by relevance models. We apply our method to six topic sets from test collections in five different genres. Our parsimonious relevance models (i) improve retrieval effectiveness in terms of MAP on all collections, (ii) significantly outperform their non-parsimonious counterparts on most measures, and (iii) have a precision enhancing effect, unlike other blind relevance feedback methods.
Combining Candidate and Document Models for Expert Search.
Abstract: We describe our participation in the TREC 2008 Enterprise track and detail our language modeling-based approaches. For document search, our focus was on query expansion using profiles of top ranked experts and on document priors. We found that these techniques result in small, but noticeable improvements over our baseline method. For expert search, we combine candidate- and document-based models, and also bring in web evidence. We found that the combined models significantly and consistently outperformed our very competitive baseline models. 1
Resolving Person Names in Web People Search.

People search in the enterprise.

The SIGIR 2008 workshop on future challenges in expertise retrieval %28fCHER%29.
Abstract: We describe our participation in the TREC 2008 Enterprise track and detail our language modeling-based approaches. For document search, our focus was on query expansion using profiles of top ranked experts and on document priors. We found that these techniques result in small, but noticeable improvements over our baseline method. For expert search, we combine candidate- and document-based models, and also bring in web evidence. We found that the combined models significantly and consistently outperformed our very competitive baseline models. 1
Determining Expert Profiles %28With an Application to Expert Finding%29.

Building simulated queries for known%2Ditem topics an analysis using six european languages.

Broad expertise retrieval in sparse data environments.
Expertise retrieval has been largely unexplored on data other than the W3C collection. At the same time, many intranets of universities and other knowledge-intensive organisations offer examples of relatively small but clean multilingual expertise data, covering broad ranges of expertise areas. We first present two main expertise retrieval tasks, along with a set of baseline approaches based on generative language modeling, aimed at finding expertise relations between topics and people. For our experimental evaluation, we introduce (and release) a new test set based on a crawl of a university site. Using this test set, we conduct two series of experiments. The first is aimed at determining the effectiveness of baseline expertise retrieval methods applied to the new test set. The second is aimed at assessing refined models that exploit characteristic features of the new test set, such as the organizational structure of the university, and the hierarchical structure of the topics in the test set. Expertise retrieval models are shown to be robust with respect to environments smaller than the W3C collection, and current techniques appear to be generalizable to other settings.
Finding similar experts.
Expert finding has been addressed from multiple viewpoints, including expertise seeking and expert retrieval. The focus of expertise seeking has mostly been on descriptive or predictive models, for example to identify what factors affect human decisions on locating and selecting experts. In expert retrieval the focus has been on algorithms similar to document search, which identify topical matches based on the content of documents associated with experts. We report on a pilot study on an expert finding task in which we explore how contextual factors identified by expertise seeking models can be integrated with topic-centric retrieval algorithms and examine whether they can improve retrieval performance for this task. We focus on the task of similar expert finding: given a small number of example experts, find similar experts. Our main finding is that, while topical knowledge is the most important factor, human subjects also consider other factors, such as reliability, upto-dateness, and organizational structure. We find that integrating these factors into topical retrieval models can significantly improve retrieval performance.
People search in the enterprise.

Query and Document Models for Enterprise Search.
Abstract: We describe our participation in the TREC 2007 Enterprise track and detail our language modeling-based approaches. For document search, our focus was on estimating a mixture model using a standard web collection, and on constructing query models by employing blind relevance feedback and using the example documents provided with the topics. We found that settings performing well on a web collection do not carry over to the CSIRO collection, but the use of advanced query models resulted in significant improvements. In expert search, our experiments concerned document representation, identification of candidate experts, and combinations of expert search strategies. We find no significant difference in average precision but observe small overall positive effects of the advanced models, with large differences between individual topics. 1

Overview of WebCLEF 2006.
Abstract. We report on the CLEF 2006 WebCLEF track devoted to crosslingual web retrieval. We provide details about the retrieval tasks, the used topic set, and the results of the participants. WebCLEF 2006 used a stream of known-item topics consisting of: (i) manual topics (including a selection of WebCLEF 2005 topics, and a set of new topics) and (ii) automatically generated topics (generated using two techniques). The results over all topics show that current CLIR systems are very effective, retrieving on average the target page in the top ranks. Manually constructed topics result in higher performance than and automatically generated ones. And finally, the resulting scores on automatic topics provide a reasonable ranking of the systems, showing that automatically generated topics are an attractive alternative in situations where manual topics are not readily available. 1

Why Are They Excited%3F Identifying and Explaining Spikes in Blog Mood Levels.
We describe a method for discovering irregularities in temporal mood patterns appearing in a large corpus of blog posts, and labeling them with a natural language explanation. Simple techniques based on comparing corpus frequencies, coupled with large quantities of data, are shown to be effective for identifying the events underlying changes in global moods. 1
Formal models for expert finding in enterprise corpora.

Language Models for Enterprise Search Query Expansion and Combination of Evidence.
 We describe our participation in the TREC 2006 Enterprise track. We provide a detailed account of the ideas underlying our language modeling approaches to both the discussion search and expert search tasks. For discussion search, our focus was on query expansion techniques, using additional information from the topic statement and from message threads; while the former was generally helpful, the latter mostly hurt performance. In expert search our main experiments concerned query expansion as well as combinations of expert finding and expert profiling techniques.   
Finding experts and their eetails in e%2Dmail corpora.

Language Modeling Approaches for Enterprise Tasks.
We describe our participation in the  TREC 2005 Enterprise track. We provide a detailed  account of the ideas underlying our language  modeling approaches to these tasks, report  on our results, and give a summary of our findings  so far.
