Named entity recognition studies the problem of locating and classifying parts of free text into a set of predefined categories. Although extensive research has focused on the detection of person, location and organization entities, there are many other entities of interest, including phone numbers, dates, times and currencies (to name a few examples). We refer to these types of entities as "semi-structured named entities", since they usually follow certain syntactic formats according to some conventions, although their structure is typically not well-defined. Regular expression solutions require significant amount of manual effort and supervised machine learning approaches rely on large sets of labeled training data. Therefore, these approaches do not scale when we need to support many semi-structured entity types in many languages and regions.

The problem of automatically extracting the most interesting and relevant keyword phrases in a document has been studied extensively as it is crucial for a number of applications. These applications include contextual advertising, automatic text summarization, and user-centric entity detection systems. All these applications can potentially benefit from a successful solution as it enables computational efficiency (by decreasing the input size), noise reduction, or overall improved user satisfaction.In this paper, we study this problem and focus on improving the overall quality of user-centric entity detection systems. First, we review our concept extraction technique, which relies on search engine query logs. We then define a new feature space to represent the interesting ness of concepts, and describe a new approach to estimate their relevancy for a given context. We utilize click through data obtained from a large scale user-centric entity detection system - Contextual Shortcuts - to train a model to rank the extracted concepts, and evaluate the resulting model extensively again based on their click through data. Our results show that the learned model outperforms the baseline model, which employs similar features but whose weights are tuned carefully based on empirical observations, and reduces the error rate from 30.22% to 18.66%.

A user-centric entity detection system is one in which the primary consumer of the detected entities is a person who can perform actions on the detected entities (e.g. perform a search, view a map, shop, etc.). We contrast this with machine-centric detection systems where the primary consumer of the detected entities is a machine. Machine-centric detection systems typically focus on the quantity of detected entities, measured by precision and recall metrics, with the goal of correctly identifying every single entity in a document.

Contextual search refers to proactively capturing the information need of a user by automatically augmenting the user query with information extracted from the search context; for example, by using terms from the web page the user is currently browsing or a file the user is currently editing.We present three different algorithms to implement contextual search for the Web. The first, it query rewriting (QR), augments each query with appropriate terms from the search context and uses an off-the-shelf web search engine to answer this augmented query. The second, rank-biasing (RB), generates a representation of the context and answers queries using a custom-built search engine that exploits this representation. The third, iterative filtering meta-search (IFM), generates multiple subqueries based on the user query and appropriate terms from the search context, uses an off-the-shelf search engine to answer these subqueries, and re-ranks the results of the subqueries using rank aggregation methods.We extensively evaluate the three methods using 200 contexts and over 24,000 human relevance judgments of search results. We show that while QR works surprisingly well, the relevance and recall can be improved using RB and substantially more using IFM. Thus, QR, RB, and IFM represent a cost-effective design spectrum for contextual search.

Contextual search tries to better capture a user's information need by augmenting the user's query with contextual information extracted from the search context (for example, terms from the web page the user is currently reading or a file the user is currently editing).This paper presents Y!Q---a first of its kind large-scale contextual search system---and provides an overview of its system design and architecture. Y!Q solves two major problems. First, how to capture high quality search context. Second, how to use that context in a way to improve the relevancy of search queries. To address the first problem, Y!Q introduces an information widget that captures precise search context and provides convenient access to its functionality at the point of inspiration. For example, Y!Q can be easily embedded into web pages using a web API, or it can be integrated into a web browser toolbar. This paper provides an overview of Y!Q's user interaction design, highlighting its novel aspects for capturing high quality search context.To address the second problem, Y!Q uses a semantic network for analyzing search context, possibly resolving ambiguous terms, and generating a contextual digest comprising its key concepts. This digest is passed through a query planner and rewriting framework for augmenting a user's search query with relevant context terms to improve the overall search relevancy and experience. We show experimental results comparing contextual Y!Q search results side-by-side with regular Yahoo! web search results. This evaluation suggests that Y!Q results are considered significantly more relevant.The paper also identifies interesting research problems and argues that contextual search may represent the next major step in the evolution of web search engines.

When searching large hypertext document collections, it is often possible that there are too many results available for ambiguous queries. Query refinement is an interactive process of query modification that can be used to narrow down the scope of search results. We propose a new method for automatically generating refinements or related terms to queries by mining anchor text for a large hypertext document collection. We show that the usage of anchor text as a basis for query refinement produces high quality refinement suggestions that are significantly better in terms of perceived usefulness compared to refinements that are derived using the document content. Furthermore, our study suggests that anchor text refinements can also be used to augment traditional query refinement algorithms based on query logs, since they typically differ in coverage and produce different refinements. Our results are based on experiments on an anchor text collection of a large corporate intranet.