Queries asked on search engines nowa-days increasingly fall in full natural lan-guage, which refer to Natural Language queries (NL queries). Parsing that kind of queries for the purpose of understanding user’s query intent is an essential factor to search engine. To this end, a hierarchical structure is introduced to represent the se-mantic intent of NL query and then we fo-cus on the problem of mapping NL queries to the corresponding semantic intents. We propose a parsing method by conducting two steps as follows: (1) predicting se-mantic tags for a given input query; (2) building an intent representation for the query using the sequence of semantic tags based on a structured SVM classification model. Experimental results on a manu-ally labeled corpus show that our method achieved a sufficiently high result in term of precision and F1. 1

Proposed duration is a full-day tutorial. The current plan is to di-vide the tutorial into two main parts, each focusing on applications of the discussed techniques to verbose natural language queries. 1. Query reduction, reformulation and segmentation techniques. 2. Query concept weighting, expansion and learning-to-rank. Contact Information

We present a generic method for augmenting unsupervised query segmentation by incorporating Parts-of-Speech (POS) sequence information to detect meaningful but rare n-grams. Our initial experiments with an existing English POS tagger employing two different POS tagsets and an unsupervised POS induction technique specifically adapted for queries show that POS information can significantly improve query segmentation performance in all these cases.

Named Entity Recognition and Classification is the task of extracting from text, instances of different entity classes such as person, location, or company. This task has recently been applied to web search queries in order to better understand their semantics, where a search query consists of linguistic units that users submit to a search engine to convey their search need. Discovering and analysing the linguistic units comprising a search query enables search engines to reveal and meet users ’ search intents. As a result, recent research has concentrated on analysing the constituent units comprising search queries. However, since search queries are short, unstructured, and ambiguous, an approach to detect and classify named entities is presented in this thesis, in which queries are augmented with the text snippets of search results for search queries. The thesis makes the following contributions: 1. A novel method for detecting candidate named entities in search queries, which utilises both query grammatical annotation and query segmentation. 2. A novel method to classify the detected candidate entities into a set of target entity classes, by using a seed expansion approach; the method presented exploits the represen-tation of the sets of contextual clues surrounding the entities in the snippets as vectors in a common vector space. 3. An exploratory analysis of three main categories of search refiners: nouns, verbs, and adjectives, that users often incorporate in entity-centric queries in order to further refine the entity-related search results. 4. A taxonomy of named entities derived from a search engine query log. By using a large commercial query log, experimental evidence is provided that the work presented herein is competitive with the existing research in the field of entity recognition and classification in search queries.

Web users are increasingly looking for structured data, such as lyrics, job, or recipes, using unstructured queries on the web. However, retrieving relevant results from such data is a challenging problem due to the unstructured language of the web queries. In this paper, we propose a method to improve web search ranking by detecting Structured Annotation of queries based on top search results. In a structured annotation, the original query is split into different units that are associated with semantic attributes in the corresponding domain. We evaluate our techniques using real world queries and achieve significant improvement. 1

We review the state-of-the-art in deep web search and propose a novel classification scheme to better compare deep web search systems. The current binary classification (surfacing versus virtual integration) hides a number of implicit decisions that must be made by a developer. We make these decisions explicit by distinguishing 7 system aspects that describe a system in terms of its functionality (what it can, and what it cannot do) and in terms of its solution to a specific problem. We then motivate the need for a search system which has a singlefield free-text query interface that supports real-time structured search over multiple sources. To this end, we discuss two possible federated architectures and state the scientific challenges. Finally, we present the findings of our ongoing project and briefly outline related work to freetext interfaces over structured data.