Search (40 results, page 1 of 2)

0.02441524 = product of:
  0.09766096 = sum of:
    0.09766096 = weight(_text_:social in 4097) [ClassicSimilarity], result of:
      0.09766096 = score(doc=4097,freq=8.0), product of:
        0.1847249 = queryWeight, product of:
          3.9875789 = idf(docFreq=2228, maxDocs=44218)
          0.046325076 = queryNorm
        0.52868325 = fieldWeight in 4097, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          3.9875789 = idf(docFreq=2228, maxDocs=44218)
          0.046875 = fieldNorm(doc=4097)
  0.25 = coord(1/4)

Abstract

A challenge of managing and extracting useful knowledge from social media data sources has attracted much attention from academic and industry. To address this challenge, semantic analysis of textual data is focused in this paper. We propose an ontology-based approach to extract semantics of textual data and define the domain of data. In other words, we semantically analyse the social data at two levels i.e. the entity level and the domain level. We have chosen Twitter as a social channel challenge for a purpose of concept proof. Domain knowledge is captured in ontologies which are then used to enrich the semantics of tweets provided with specific semantic conceptual representation of entities that appear in the tweets. Case studies are used to demonstrate this approach. We experiment and evaluate our proposed approach with a public dataset collected from Twitter and from the politics domain. The ontology-based approach leverages entity extraction and concept mappings in terms of quantity and accuracy of concept identification.

0.017264182 = product of:
  0.06905673 = sum of:
    0.06905673 = weight(_text_:social in 3374) [ClassicSimilarity], result of:
      0.06905673 = score(doc=3374,freq=4.0), product of:
        0.1847249 = queryWeight, product of:
          3.9875789 = idf(docFreq=2228, maxDocs=44218)
          0.046325076 = queryNorm
        0.3738355 = fieldWeight in 3374, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.9875789 = idf(docFreq=2228, maxDocs=44218)
          0.046875 = fieldNorm(doc=3374)
  0.25 = coord(1/4)

Abstract

As the amount of content and the number of users in social relationships is continually growing in the Internet, resource sharing and access policy management is difficult, time-consuming and error-prone. Cross-domain recommendation of private or protected resources managed and secured by each domain's specific access rules is impracticable due to private security policies and poor sharing mechanisms. This work focus on exploiting resource's content, user's preferences, users' social networks and semantic information to cross-relate different resources through their meta information using recommendation techniques that combine collaborative-filtering techniques with semantics annotations, by generating associations between resources. The semantic similarities established between resources are used on a hybrid recommendation engine that interprets user and resources' semantic information. The recommendation engine allows the promotion and discovery of unknown-unknown resources to users that could not even know about the existence of those resources thus providing means to solve the cross-domain recommendation of private or protected resources.

0.012938853 = product of:
  0.051755413 = sum of:
    0.051755413 = product of:
      0.10351083 = sum of:
        0.10351083 = weight(_text_:aspects in 4704) [ClassicSimilarity], result of:
          0.10351083 = score(doc=4704,freq=4.0), product of:
            0.20938325 = queryWeight, product of:
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.046325076 = queryNorm
            0.4943606 = fieldWeight in 4704, product of:
              2.0 = tf(freq=4.0), with freq of:
                4.0 = termFreq=4.0
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.0546875 = fieldNorm(doc=4704)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

Focuses on the user and human-computer interaction (HCI) aspects of the research based on the Okapi text retrieval system. Describes 3 experiments using different approaches to query expansion, highlighting the relationship between the functionality of a system and different interface designs. These experiments involve both automatic and interactive query expansion, and both character based and GUI (graphical user interface) environments. The effectiveness of the search interaction for query expansion depends on resolving opposing interface and functional aspects, e.g. automatic vs. interactive query expansion, explicit vs. implicit use of a thesaurus, and document vs. query space

0.012938853 = product of:
  0.051755413 = sum of:
    0.051755413 = product of:
      0.10351083 = sum of:
        0.10351083 = weight(_text_:aspects in 5166) [ClassicSimilarity], result of:
          0.10351083 = score(doc=5166,freq=4.0), product of:
            0.20938325 = queryWeight, product of:
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.046325076 = queryNorm
            0.4943606 = fieldWeight in 5166, product of:
              2.0 = tf(freq=4.0), with freq of:
                4.0 = termFreq=4.0
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.0546875 = fieldNorm(doc=5166)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

Knowledge organization systems have been studied in several fields and for different and complementary aspects. Among the aspects that concentrate common interests, in this article we highlight those related to the terminological and conceptual relationships among the components of any knowledge organization system. This research aims to contribute to the critical analysis of knowledge organization systems, especially ontologies, thesauri, and classification systems, by the comprehension of its similarities and differences when dealing with concepts and their ways of relating to each other as well as to the conceptual design that is adopted.

0.011095227 = product of:
  0.044380907 = sum of:
    0.044380907 = product of:
      0.088761814 = sum of:
        0.088761814 = weight(_text_:22 in 1352) [ClassicSimilarity], result of:
          0.088761814 = score(doc=1352,freq=4.0), product of:
            0.16222252 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.046325076 = queryNorm
            0.54716086 = fieldWeight in 1352, product of:
              2.0 = tf(freq=4.0), with freq of:
                4.0 = termFreq=4.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.078125 = fieldNorm(doc=1352)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Date

22. 2.2003 17:25:39
22. 2.2003 18:17:40

0.010983714 = product of:
  0.043934856 = sum of:
    0.043934856 = product of:
      0.08786971 = sum of:
        0.08786971 = weight(_text_:22 in 2134) [ClassicSimilarity], result of:
          0.08786971 = score(doc=2134,freq=2.0), product of:
            0.16222252 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.046325076 = queryNorm
            0.5416616 = fieldWeight in 2134, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.109375 = fieldNorm(doc=2134)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Date

30. 3.2001 13:32:22

0.010173016 = product of:
  0.040692065 = sum of:
    0.040692065 = weight(_text_:social in 2588) [ClassicSimilarity], result of:
      0.040692065 = score(doc=2588,freq=2.0), product of:
        0.1847249 = queryWeight, product of:
          3.9875789 = idf(docFreq=2228, maxDocs=44218)
          0.046325076 = queryNorm
        0.22028469 = fieldWeight in 2588, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.9875789 = idf(docFreq=2228, maxDocs=44218)
          0.0390625 = fieldNorm(doc=2588)
  0.25 = coord(1/4)

Abstract

In human-computer interaction (HCI), a successful interaction sequence can take its own momentum and drift away from what the user has originally planned. However, this does not mean that planned actions play no important role in the overall performance. In this paper, the author tries to construct a line of argument to demonstrate that it is impossible to consider an action without an a priori plan, even according to the phenomenological position taken for granted by the situated action theory. Based on the phenomenological analysis of problematic situations and typification the author argues that, just like "situated-ness", "planned-ness" of an action should also be understood in the context of the situation. Successful plan can be developed and executed for familiar context. The first part of the paper treats information seeking behavior as a special type of social action and applies Alfred Schutz's phenomenology of sociology to understand the importance and necessity of plan. The second part reports results of a quasi-experiment focusing on plan deviation within an information seeking context. It was found that when the searcher's situation changed from problematic to non-problematic, the degree of plan deviation decreased significantly. These results support the argument proposed in the first part of the paper.

0.009242038 = product of:
  0.036968153 = sum of:
    0.036968153 = product of:
      0.073936306 = sum of:
        0.073936306 = weight(_text_:aspects in 624) [ClassicSimilarity], result of:
          0.073936306 = score(doc=624,freq=4.0), product of:
            0.20938325 = queryWeight, product of:
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.046325076 = queryNorm
            0.35311472 = fieldWeight in 624, product of:
              2.0 = tf(freq=4.0), with freq of:
                4.0 = termFreq=4.0
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.0390625 = fieldNorm(doc=624)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

With RetrievalWare 8.0(TM) the American company Convera offers an elaborated software in the range of Information Retrieval, Information Indexing and Knowledge Management. Convera promises the possibility of handling different file formats in many different languages. Regarding comparable products one innovation is to be stressed particularly: the possibility of the preparation as well as integration of an ontology. One tool of the software package is useful in order to produce ontologies manually, to process existing ontologies and to import the very. The processing of search results is also to be mentioned. By means of categorization strategies search results can be classified dynamically and presented in personalized representations. This study presents an evaluation of the functions and components of the system. Technological aspects and modes of operation under the surface of Convera RetrievalWare will be analysed, with a focus on the creation of libraries and thesauri, and the problems posed by the integration of an existing thesaurus. Broader aspects such as usability and system ergonomics are integrated in the examination as well.

0.008138414 = product of:
  0.032553654 = sum of:
    0.032553654 = weight(_text_:social in 276) [ClassicSimilarity], result of:
      0.032553654 = score(doc=276,freq=2.0), product of:
        0.1847249 = queryWeight, product of:
          3.9875789 = idf(docFreq=2228, maxDocs=44218)
          0.046325076 = queryNorm
        0.17622775 = fieldWeight in 276, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.9875789 = idf(docFreq=2228, maxDocs=44218)
          0.03125 = fieldNorm(doc=276)
  0.25 = coord(1/4)

Abstract

Context plays an important role in a number of domains where reasoning intervenes as in understanding, interpretation, diagnosis, etc. The reason is that reasoning activities heavily rely on a background (or experience) that is generally not made explicit and that gives a contextual dimension to knowledge. On the Web in December 1996, AItaVista gave more than 710000 pages containing the word context, when concept gives only 639000 references. A clear definition of this word stays to be found. There are several formal definitions of this concept (references are given in Brézillon, 1996): a set of preferences and/or beliefs, an infinite and only partially known collection of assumptions, a list of attributes, the product of an interpretation, possible worlds, assumptions under which a statement is true or false. One faces the same situation at the programming level: a collection of context schemas; a path in information retrieval; slots in object-oriented languages; a special, buffer-like data structure; a window on the screen, buttons which are functional customisable and shareable; an interpreter which controls the system's activity; the characteristics of the situation and the goals of the knowledge use; or entities (things or events) related in a certain way that permits to listen what is said and what is not said. Context is often assimilated at a set of restrictions (e.g., preconditions) that limit access to parts of the applications. The first works considering context explicitly are in Natural Language. Researchers in this domain focus on the linguistic context, sometimes associated with other types of contexts as: semantic context, cognitive context, physical and perceptual context, and social context (Bunt, 1997).

0.007845511 = product of:
  0.031382043 = sum of:
    0.031382043 = product of:
      0.062764086 = sum of:
        0.062764086 = weight(_text_:22 in 2751) [ClassicSimilarity], result of:
          0.062764086 = score(doc=2751,freq=2.0), product of:
            0.16222252 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.046325076 = queryNorm
            0.38690117 = fieldWeight in 2751, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.078125 = fieldNorm(doc=2751)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Date

1. 2.2016 18:25:22

0.007845511 = product of:
  0.031382043 = sum of:
    0.031382043 = product of:
      0.062764086 = sum of:
        0.062764086 = weight(_text_:22 in 2754) [ClassicSimilarity], result of:
          0.062764086 = score(doc=2754,freq=2.0), product of:
            0.16222252 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.046325076 = queryNorm
            0.38690117 = fieldWeight in 2754, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.078125 = fieldNorm(doc=2754)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Date

1. 2.2016 18:25:22

0.007845511 = product of:
  0.031382043 = sum of:
    0.031382043 = product of:
      0.062764086 = sum of:
        0.062764086 = weight(_text_:22 in 3279) [ClassicSimilarity], result of:
          0.062764086 = score(doc=3279,freq=2.0), product of:
            0.16222252 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.046325076 = queryNorm
            0.38690117 = fieldWeight in 3279, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.078125 = fieldNorm(doc=3279)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Source

Metadata and semantics research: 10th International Conference, MTSR 2016, Göttingen, Germany, November 22-25, 2016, Proceedings. Eds.: E. Garoufallou

0.007845511 = product of:
  0.031382043 = sum of:
    0.031382043 = product of:
      0.062764086 = sum of:
        0.062764086 = weight(_text_:22 in 3280) [ClassicSimilarity], result of:
          0.062764086 = score(doc=3280,freq=2.0), product of:
            0.16222252 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.046325076 = queryNorm
            0.38690117 = fieldWeight in 3280, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.078125 = fieldNorm(doc=3280)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Source

Metadata and semantics research: 10th International Conference, MTSR 2016, Göttingen, Germany, November 22-25, 2016, Proceedings. Eds.: E. Garoufallou

0.007842129 = product of:
  0.031368516 = sum of:
    0.031368516 = product of:
      0.06273703 = sum of:
        0.06273703 = weight(_text_:aspects in 2091) [ClassicSimilarity], result of:
          0.06273703 = score(doc=2091,freq=2.0), product of:
            0.20938325 = queryWeight, product of:
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.046325076 = queryNorm
            0.29962775 = fieldWeight in 2091, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.046875 = fieldNorm(doc=2091)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

The ever growing popularity of the Internet as a source of information, coupled with the accompanying growth in the number of documents made available through the World Wide Web, is leading to an increasing demand for more efficient and accurate information retrieval tools. Numerous techniques have been proposed and tried for improving the effectiveness of searching the World Wide Web for documents relevant to a given topic of interest. The specification of appropriate keywords and phrases by the user is crucial for the successful execution of a query as measured by the relevance of documents retrieved. Lack of users' knowledge an the search topic and their changing information needs often make it difficult for them to find suitable keywords or phrases for a query. This results in searches that fail to cover all likely aspects of the topic of interest. We describe a scheme that attempts to remedy this situation by automatically expanding the user query through the analysis of initially retrieved documents. Experimental results to demonstrate the effectiveness of the query expansion scheure are presented.

0.007842129 = product of:
  0.031368516 = sum of:
    0.031368516 = product of:
      0.06273703 = sum of:
        0.06273703 = weight(_text_:aspects in 3823) [ClassicSimilarity], result of:
          0.06273703 = score(doc=3823,freq=2.0), product of:
            0.20938325 = queryWeight, product of:
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.046325076 = queryNorm
            0.29962775 = fieldWeight in 3823, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.046875 = fieldNorm(doc=3823)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

Online trolling has become increasingly prevalent and visible in online communities. Perceptions of and reactions to trolling behaviors varies significantly from one community to another, as trolling behaviors are contextual and vary across platforms and communities. Through an examination of seven trolling scenarios, this article intends to answer the following questions: how do trolling behaviors differ across contexts; how do perceptions of trolling differ from case to case; and what aspects of context of trolling are perceived to be important by the public? Based on focus groups and interview data, we discuss the ways in which community norms and demographics, technological features of platforms, and community boundaries are perceived to impact trolling behaviors. Two major contributions of the study include a codebook to support future analysis of trolling and formal concept analysis surrounding contextual perceptions of trolling.

0.007766659 = product of:
  0.031066636 = sum of:
    0.031066636 = product of:
      0.06213327 = sum of:
        0.06213327 = weight(_text_:22 in 5295) [ClassicSimilarity], result of:
          0.06213327 = score(doc=5295,freq=4.0), product of:
            0.16222252 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.046325076 = queryNorm
            0.38301262 = fieldWeight in 5295, product of:
              2.0 = tf(freq=4.0), with freq of:
                4.0 = termFreq=4.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.0546875 = fieldNorm(doc=5295)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Date

22. 7.2006 17:56:22

0.0065351077 = product of:
  0.026140431 = sum of:
    0.026140431 = product of:
      0.052280862 = sum of:
        0.052280862 = weight(_text_:aspects in 206) [ClassicSimilarity], result of:
          0.052280862 = score(doc=206,freq=2.0), product of:
            0.20938325 = queryWeight, product of:
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.046325076 = queryNorm
            0.2496898 = fieldWeight in 206, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.0390625 = fieldNorm(doc=206)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

Context is one of the most important concepts in information seeking and retrieval research. However, the challenges of studying context are great; thus, it is more common for researchers to use context as a post hoc explanatory factor, rather than as a concept that drives inquiry. The purposes of this study were to develop a method for collecting data about information seeking context in natural online environments, and identify which aspects of context should be considered when studying online information seeking. The study is reported in two parts. In this, the first part, the background and method are presented. Results and implications of this research are presented in Part 2 (Kelly, in press). Part 1 discusses previous literature on information seeking context and behavior and situates the current work within this literature. This part further describes the naturalistic, longitudinal research design that was used to examine and measure the online information seeking contexts of users during a 14-week period. In this design, information seeking context was characterized by a user's self-identified tasks and topics, and several attributes of these, such as the length of time the user expected to work on a task and the user's familiarity with a topic. At weekly intervals, users evaluated the usefulness of the documents that they viewed, and classified these documents according to their tasks and topics. At the end of the study, users provided feedback about the study method.

0.0065351077 = product of:
  0.026140431 = sum of:
    0.026140431 = product of:
      0.052280862 = sum of:
        0.052280862 = weight(_text_:aspects in 215) [ClassicSimilarity], result of:
          0.052280862 = score(doc=215,freq=2.0), product of:
            0.20938325 = queryWeight, product of:
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.046325076 = queryNorm
            0.2496898 = fieldWeight in 215, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.0390625 = fieldNorm(doc=215)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

Context is one of the most important concepts in information seeking and retrieval research. However, the challenges of studying context are great; thus, it is more common for researchers to use context as a post hoc explanatory factor, rather than as a concept that drives inquiry. The purpose of this study was to develop a method for collecting data about information seeking context in natural online environments, and identify which aspects of context should be considered when studying online information seeking. The study is reported in two parts. In this, the second part, results and implications of this research are presented. Part 1 (Kelly, 2006) discussed previous literature on information seeking context and behavior, situated the current study within this literature, and described the naturalistic, longitudinal research design that was used to examine and measure the online information seeking context of seven users during a 14-week period. Results provide support for the value of the method in studying online information seeking context, the relative importance of various measures of context, how these measures change over time, and, finally, the relationship between these measures. In particular, results demonstrate significant differences in distributions of usefulness ratings according to task and topic.

0.0065351077 = product of:
  0.026140431 = sum of:
    0.026140431 = product of:
      0.052280862 = sum of:
        0.052280862 = weight(_text_:aspects in 1105) [ClassicSimilarity], result of:
          0.052280862 = score(doc=1105,freq=2.0), product of:
            0.20938325 = queryWeight, product of:
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.046325076 = queryNorm
            0.2496898 = fieldWeight in 1105, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.0390625 = fieldNorm(doc=1105)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

Topic ontologies or web directories consist of large collections of links to websites, arranged by topic in different categories. The structure of these ontologies is typically not flat because there are hierarchical and nonhierarchical relationships among topics. As a consequence, websites classified under a certain topic may be relevant to other topics. Although some of these relevance relations are explicit, most of them must be discovered by an analysis of the structure of the ontologies. This article proposes a family of models of relevance propagation in topic ontologies. An efficient computational framework is described and used to compute nine different models for a portion of the Open Directory Project graph consisting of more than half a million nodes and approximately 1.5 million edges of different types. After performing a quantitative analysis, a user study was carried out to compare the most promising models. It was found that some general difficulties rule out the possibility of defining flawless models of relevance propagation that only take into account structural aspects of an ontology. However, there is a clear indication that including transitive relations induced by the nonhierarchical components of the ontology results in relevance propagation models that are superior to more basic approaches.

0.0065351077 = product of:
  0.026140431 = sum of:
    0.026140431 = product of:
      0.052280862 = sum of:
        0.052280862 = weight(_text_:aspects in 1545) [ClassicSimilarity], result of:
          0.052280862 = score(doc=1545,freq=2.0), product of:
            0.20938325 = queryWeight, product of:
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.046325076 = queryNorm
            0.2496898 = fieldWeight in 1545, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              4.5198684 = idf(docFreq=1308, maxDocs=44218)
              0.0390625 = fieldNorm(doc=1545)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

This study proposes a new way of using WordNet for query expansion (QE). We choose candidate expansion terms from a set of pseudo-relevant documents; however, the usefulness of these terms is measured based on their definitions provided in a hand-crafted lexical resource such as WordNet. Experiments with a number of standard TREC collections WordNet-based that this method outperforms existing WordNet-based methods. It also compares favorably with established QE methods such as KLD and RM3. Leveraging earlier work in which a combination of QE methods was found to outperform each individual method (as well as other well-known QE methods), we next propose a combination-based QE method that takes into account three different aspects of a candidate expansion term's usefulness: (a) its distribution in the pseudo-relevant documents and in the target corpus, (b) its statistical association with query terms, and (c) its semantic relation with the query, as determined by the overlap between the WordNet definitions of the term and query terms. This combination of diverse sources of information appears to work well on a number of test collections, viz., TREC123, TREC5, TREC678, TREC robust (new), and TREC910 collections, and yields significant improvements over competing methods on most of these collections.