Search (221 results, page 3 of 12)

0.0023499418 = product of:
  0.0046998835 = sum of:
    0.0046998835 = product of:
      0.009399767 = sum of:
        0.009399767 = weight(_text_:a in 1704) [ClassicSimilarity], result of:
          0.009399767 = score(doc=1704,freq=16.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.2161963 = fieldWeight in 1704, product of:
              4.0 = tf(freq=16.0), with freq of:
                16.0 = termFreq=16.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=1704)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

A procedure that processes a corpus of text and produces numeric vectors containing information about its meanings for each word is presented. This procedure is applied to a large corpus of natural language text taken from Usenet, and the resulting vectors are examined to determine what information is contained within them. These vectors provide the coordinates in a high-dimensional space in which word relationships can be analyzed. Analyses of both vector similarity and multidimensional scaling demonstrate that there is significant semantic information carried in the vectors. A comparison of vector similarity with human reaction times in a single-word priming experiment is presented. These vectors provide the basis for a representational model of semantic memory, hyperspace analogue to language (HAL).

Type

a

0.0023499418 = product of:
  0.0046998835 = sum of:
    0.0046998835 = product of:
      0.009399767 = sum of:
        0.009399767 = weight(_text_:a in 2152) [ClassicSimilarity], result of:
          0.009399767 = score(doc=2152,freq=16.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.2161963 = fieldWeight in 2152, product of:
              4.0 = tf(freq=16.0), with freq of:
                16.0 = termFreq=16.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=2152)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

In this paper, we compare a well-known semantic spacemodel, Latent Semantic Analysis (LSA) with another model, Hyperspace Analogue to Language (HAL) which is widely used in different area, especially in automatic query refinement. We conduct this comparative analysis to prove our hypothesis that with respect to ability of extracting the lexical information from a corpus of text, LSA is quite similar to HAL. We regard HAL and LSA as black boxes. Through a Pearson's correlation analysis to the outputs of these two black boxes, we conclude that LSA highly co-relates with HAL and thus there is a justification that LSA and HAL can potentially play a similar role in the area of facilitating automatic query refinement. This paper evaluates LSA in a new application area and contributes an effective way to compare different semantic space models.

Type

a

0.0023499418 = product of:
  0.0046998835 = sum of:
    0.0046998835 = product of:
      0.009399767 = sum of:
        0.009399767 = weight(_text_:a in 2259) [ClassicSimilarity], result of:
          0.009399767 = score(doc=2259,freq=16.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.2161963 = fieldWeight in 2259, product of:
              4.0 = tf(freq=16.0), with freq of:
                16.0 = termFreq=16.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=2259)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

An ontology is "an explicit conceptualization of a domain of discourse, and thus provides a shared and common understanding of the domain." We have been producing ontologies for millennia to understand and explain our rationale and environment. From Plato's philosophical framework to modern day classification systems, ontologies are, in most cases, the product of extensive analysis and categorization. Only recently has the process of building ontologies become a research topic of interest. Today, ontologies are built very much ad-hoc. A terminology is first developed providing a controlled vocabulary for the subject area or domain of interest, then it is organized into a taxonomy where key concepts are identified, and finally these concepts are defined and related to create an ontology. The intent of this paper is to show that domain analysis methods can be used for building ontologies. Domain analysis aims at generic models that represent groups of similar systems within an application domain. In this sense, it deals with categorization of common objects and operations, with clear, unambiguous definitions of them and with defining their relationships.

Type

a

0.0023499418 = product of:
  0.0046998835 = sum of:
    0.0046998835 = product of:
      0.009399767 = sum of:
        0.009399767 = weight(_text_:a in 2384) [ClassicSimilarity], result of:
          0.009399767 = score(doc=2384,freq=16.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.2161963 = fieldWeight in 2384, product of:
              4.0 = tf(freq=16.0), with freq of:
                16.0 = termFreq=16.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=2384)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

LyberWorld is a prototype IR user interface. It implements visualizations of an abstract information space-fulltext. The paper derives a model for such visualizations and an exemplar user interface design is implemented for the probabilistic fulltext retrieval system INQUERY. Visualizations are used to communicate information search and browsing activities in a natural way by applying metaphors of spatial navigation in abstract information spaces. Visualization tools for exploring information spaces and judging relevance of information items are introduced and an example session demonstrates the prototype. The presence of a spatial model in the user's mind and interaction with a system's corresponding display methods is regarded as an essential contribution towards natural interaction and reduction of cognitive costs during e.g. query construction, orientation within the database content, relevance judgement and orientation within the retrieval context.

Type

a

0.0023499418 = product of:
  0.0046998835 = sum of:
    0.0046998835 = product of:
      0.009399767 = sum of:
        0.009399767 = weight(_text_:a in 3185) [ClassicSimilarity], result of:
          0.009399767 = score(doc=3185,freq=16.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.2161963 = fieldWeight in 3185, product of:
              4.0 = tf(freq=16.0), with freq of:
                16.0 = termFreq=16.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=3185)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Query suggestion is generally an integrated part of web search engines. In this study, we first redefine and reduce the query suggestion problem as "comparison of queries". We then propose a general modular framework for query suggestion algorithm development. We also develop new query suggestion algorithms which are used in our proposed framework, exploiting query, session and user features. As a case study, we use query logs of a real educational search engine that targets K-12 students in Turkey. We also exploit educational features (course, grade) in our query suggestion algorithms. We test our framework and algorithms over a set of queries by an experiment and demonstrate a 66-90% statistically significant increase in relevance of query suggestions compared to a baseline method.

Type

a

0.0022155463 = product of:
  0.0044310926 = sum of:
    0.0044310926 = product of:
      0.008862185 = sum of:
        0.008862185 = weight(_text_:a in 8453) [ClassicSimilarity], result of:
          0.008862185 = score(doc=8453,freq=8.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20383182 = fieldWeight in 8453, product of:
              2.828427 = tf(freq=8.0), with freq of:
                8.0 = termFreq=8.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0625 = fieldNorm(doc=8453)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Evaluates the retrieval effectiveness of query expansion strategies on a test collection of the medical database MEDLINE using Cornell University's SMART retrieval system. Tests 3 expansion strategies for their ability to identify appropriate MeSH terms for user queries. Compares retrieval effectiveness using the original unexpanded and the alternative expanded user queries on a collection of 75 queries and 2.334 Medline citations. Recommends query expansions using retrieval feedback for adding MeSH search terms to a user's initial query

Type

a

0.0022155463 = product of:
  0.0044310926 = sum of:
    0.0044310926 = product of:
      0.008862185 = sum of:
        0.008862185 = weight(_text_:a in 436) [ClassicSimilarity], result of:
          0.008862185 = score(doc=436,freq=8.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20383182 = fieldWeight in 436, product of:
              2.828427 = tf(freq=8.0), with freq of:
                8.0 = termFreq=8.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0625 = fieldNorm(doc=436)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

This paper discusses the use of wiki technology to provide a navigation structure for a collection of newspaper clippings. We overview the architecture of the wiki, discuss the navigation structure and pose the question: is the navigation structure an index, and if so, what type, or is it just a linkage structure or topic map. Does such a distinction really matter? Are these definitions in reality function based?

0.0021981692 = product of:
  0.0043963385 = sum of:
    0.0043963385 = product of:
      0.008792677 = sum of:
        0.008792677 = weight(_text_:a in 5695) [ClassicSimilarity], result of:
          0.008792677 = score(doc=5695,freq=14.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20223314 = fieldWeight in 5695, product of:
              3.7416575 = tf(freq=14.0), with freq of:
                14.0 = termFreq=14.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=5695)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

To provide advice for online searching about term selection and query expansion, an intermediary expert system should indicate a terminological knowledge structure. Terminological attributes could provide the foundation of a knowledge base, and knowledge acquisition could rely on knowledge base techniques coupled with statistical techniques. The strategies of expert searchers would provide 1 source of knowledge. The knowledge structure would include 3 constructs for each term: frequency data, a hedge, and a position in a classification scheme. Switching vocabularies could provide a meta-scheme and facilitate the interoperability of databases in similar subjects. To develop such knowledge structure, research should focus on terminological attributes, word and phrase disambiguation, automated text processing, and the role of thesauri and classification schemes in indexing and retrieval. It should develop techniques that combine knowledge base and statistical methods and that consider user preferences

Type

a

0.0021981692 = product of:
  0.0043963385 = sum of:
    0.0043963385 = product of:
      0.008792677 = sum of:
        0.008792677 = weight(_text_:a in 5696) [ClassicSimilarity], result of:
          0.008792677 = score(doc=5696,freq=14.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20223314 = fieldWeight in 5696, product of:
              3.7416575 = tf(freq=14.0), with freq of:
                14.0 = termFreq=14.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=5696)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Shows how probabilistic information retrieval based on document components may be implemented as a feedforward (feedbackward) artificial neural network. The network supports adaptation of connection weights as well as the growing of new edges between queries and terms based on user relevance feedback data for training, and it reflects query modification and expansion in information retrieval. A learning rule is applied that can also be viewed as supporting sequential learning using a harmonic sequence learning rate. Experimental results with 4 standard small collections and a large Wall Street Journal collection show that small query expansion levels of about 30 terms can achieve most of the gains at the low-recall high-precision region, while larger expansion levels continue to provide gains at the high-recall low-precision region of a precision recall curve

Type

a

0.0021981692 = product of:
  0.0043963385 = sum of:
    0.0043963385 = product of:
      0.008792677 = sum of:
        0.008792677 = weight(_text_:a in 5699) [ClassicSimilarity], result of:
          0.008792677 = score(doc=5699,freq=14.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20223314 = fieldWeight in 5699, product of:
              3.7416575 = tf(freq=14.0), with freq of:
                14.0 = termFreq=14.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=5699)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

The Smart information retrieval project emphazises completely automatic approaches to the understanding and retrieval of large quantities of text. The work in the TREC-2 environment continues, performing both routing and ad hoc experiments. The ad hoc work extends investigations into combining global similarities, giving an overall indication of how a document matches a query, with local similarities identifying a smaller part of the document that matches the query. The performance of ad hoc runs is good, but it is clear that full advantage of the available local information is not been taken advantage of. The routing experiments use conventional relevance feedback approaches to routing, but with a much greater degree of query expansion than was previously done. The length of a query vector is increased by a factor of 5 to 10 by adding terms found in previously seen relevant documents. This approach improves effectiveness by 30-40% over the original query

Type

a

0.0021981692 = product of:
  0.0043963385 = sum of:
    0.0043963385 = product of:
      0.008792677 = sum of:
        0.008792677 = weight(_text_:a in 919) [ClassicSimilarity], result of:
          0.008792677 = score(doc=919,freq=14.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20223314 = fieldWeight in 919, product of:
              3.7416575 = tf(freq=14.0), with freq of:
                14.0 = termFreq=14.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=919)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Document retrieval is a highly interactive process dealing with large amounts of information. Visual representations can provide both a means for managing the complexity of large information structures and an interface style well suited to interactive manipulation. The system we have designed utilizes visually displayed graphic structures and a direct manipulation interface style to supply an integrated environment for retrieval. A common visually displayed network structure is used for query, document content, and term relations. A query can be modified through direct manipulation of its visual form by incorporating terms from any other information structure the system displays. An associative thesaurus of terms and an inter-document network provide information about a document collection that can complement other retrieval aids. Visualization of these large data structures makes use of fisheye views and overview diagrams to help overcome some of the inherent difficulties of orientation and navigation in large information structures.

Type

a

0.0021981692 = product of:
  0.0043963385 = sum of:
    0.0043963385 = product of:
      0.008792677 = sum of:
        0.008792677 = weight(_text_:a in 1082) [ClassicSimilarity], result of:
          0.008792677 = score(doc=1082,freq=14.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20223314 = fieldWeight in 1082, product of:
              3.7416575 = tf(freq=14.0), with freq of:
                14.0 = termFreq=14.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=1082)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

While surprisingly little has been written about context at a meaningful level, context is central to most theoretical approaches to information seeking. In this essay I explore in more detail three senses of context. First, I look at context as equivalent to the situation in which a process is immersed. Second, I discuss contingency approaches that detail active ingredients of the situation that have specific, predictable effects. Third, I examine major frameworks for meaning systems. Then, I discuss how a deeper appreciation of context can enhance our understanding of the process of information seeking by examining two vastly different contexts in which it occurs: organizational and cancer-related, an exemplar of everyday life information seeking. This essay concludes with a discussion of the value that can be added to information seeking research and theory as a result of a deeper appreciation of context, particularly in terms of our current multi-contextual environment and individuals taking an active role in contextualizing.

Type

a

0.0021981692 = product of:
  0.0043963385 = sum of:
    0.0043963385 = product of:
      0.008792677 = sum of:
        0.008792677 = weight(_text_:a in 2263) [ClassicSimilarity], result of:
          0.008792677 = score(doc=2263,freq=14.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20223314 = fieldWeight in 2263, product of:
              3.7416575 = tf(freq=14.0), with freq of:
                14.0 = termFreq=14.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=2263)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

In this article, the use of a new term extraction method for query expansion (QE) in text retrieval is investigated. The new method expands the initial query with a structured representation made of weighted word pairs (WWP) extracted from a set of training documents (relevance feedback). Standard text retrieval systems can handle a WWP structure through custom Boolean weighted models. We experimented with both the explicit and pseudorelevance feedback schemas and compared the proposed term extraction method with others in the literature, such as KLD and RM3. Evaluations have been conducted on a number of test collections (Text REtrivel Conference [TREC]-6, -7, -8, -9, and -10). Results demonstrated that the QE method based on this new structure outperforms the baseline.

Type

a

0.0021981692 = product of:
  0.0043963385 = sum of:
    0.0043963385 = product of:
      0.008792677 = sum of:
        0.008792677 = weight(_text_:a in 3264) [ClassicSimilarity], result of:
          0.008792677 = score(doc=3264,freq=14.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20223314 = fieldWeight in 3264, product of:
              3.7416575 = tf(freq=14.0), with freq of:
                14.0 = termFreq=14.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=3264)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

We scho how the beenfits of the view-based search method, developed within the information retrieval community, can be extended with ontology-based search, developed within the Semantic Web community, and with semantic recommendations. As a proof of the concept, we have implemented an ontology-and view-based search engine and recommendations system Ontogaotr for RDF(S) repositories. Ontogator is innovative in two ways. Firstly, the RDFS.based ontologies used for annotating metadata are used in the user interface to facilitate view-based information retrieval. The views provide the user with an overview of the repositorys contents and a vocabulary for expressing search queries. Secondlyy, a semantic browsing function is provided by a recommender system. This system enriches instance level metadata by ontologies and provides the user with links to semantically related relevant resources. The semantic linkage is specified in terms of logical rules. To illustrate and discuss the ideas, a deployed application of Ontogator to a photo repository of the Helsinki University Museum is presented.

Type

a

0.0021981692 = product of:
  0.0043963385 = sum of:
    0.0043963385 = product of:
      0.008792677 = sum of:
        0.008792677 = weight(_text_:a in 644) [ClassicSimilarity], result of:
          0.008792677 = score(doc=644,freq=14.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20223314 = fieldWeight in 644, product of:
              3.7416575 = tf(freq=14.0), with freq of:
                14.0 = termFreq=14.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=644)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

While the search interfaces used by common academic digital libraries provide easy access to a wealth of peer-reviewed literature, their interfaces provide little support for exploratory browsing. When faced with a complex search task (such as one that requires knowledge discovery), exploratory browsing is an important first step in an exploratory search process. To more effectively support exploratory browsing, we have designed and implemented a novel academic digital library search interface (KLink Search) with two new features: visually linked keywords and an interactive workspace. To study the potential value of these features, we have conducted a controlled laboratory study with 32 participants, comparing KLink Search to a baseline digital library search interface modeled after that used by IEEE Xplore. Based on subjective opinions, objective performance, and behavioral data, we show the value of adding lightweight visual and interactive features to academic digital library search interfaces to support exploratory browsing.

Type

a

0.002189429 = product of:
  0.004378858 = sum of:
    0.004378858 = product of:
      0.008757716 = sum of:
        0.008757716 = weight(_text_:a in 2589) [ClassicSimilarity], result of:
          0.008757716 = score(doc=2589,freq=20.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20142901 = fieldWeight in 2589, product of:
              4.472136 = tf(freq=20.0), with freq of:
                20.0 = termFreq=20.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0390625 = fieldNorm(doc=2589)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

The emergence of general audience digital libraries (GADLs) defines a context that represents a hybrid of both "traditional" IR, using primarily bibliographic resources provided by database vendors, and "popular" IR, exemplified by public search systems available on the World Wide Web. Findings of a study investigating end-user searching and response to a GADL are reported. Data collected from a Web-based end-user survey and data logs of resource usage for a Web-based GADL were analyzed for user characteristics, patterns of access and use, and user feedback. Cross-tabulations using respondent demographics revealed several key differences in how the system was used and valued by users of different age groups. Older users valued the service more than younger users and engaged in different searching and viewing behaviors. The GADL more closely resembles traditional retrieval systems in terms of content and purpose of use, but is more similar to popular IR systems in terms of user behavior and accessibility. A model that defines the dual context of the GADL environment is derived from the data analysis and existing IR models in general and other specific contexts. The authors demonstrate the distinguishing characteristics of this IR context, and discuss implications for the development and evaluation of future GADLs to accommodate a variety of user needs and expectations.

Type

a

0.002189429 = product of:
  0.004378858 = sum of:
    0.004378858 = product of:
      0.008757716 = sum of:
        0.008757716 = weight(_text_:a in 1105) [ClassicSimilarity], result of:
          0.008757716 = score(doc=1105,freq=20.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20142901 = fieldWeight in 1105, product of:
              4.472136 = tf(freq=20.0), with freq of:
                20.0 = termFreq=20.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0390625 = fieldNorm(doc=1105)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

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.

Type

a

0.002189429 = product of:
  0.004378858 = sum of:
    0.004378858 = product of:
      0.008757716 = sum of:
        0.008757716 = weight(_text_:a in 1545) [ClassicSimilarity], result of:
          0.008757716 = score(doc=1545,freq=20.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20142901 = fieldWeight in 1545, product of:
              4.472136 = tf(freq=20.0), with freq of:
                20.0 = termFreq=20.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0390625 = fieldNorm(doc=1545)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

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.

Type

a

0.002189429 = product of:
  0.004378858 = sum of:
    0.004378858 = product of:
      0.008757716 = sum of:
        0.008757716 = weight(_text_:a in 2877) [ClassicSimilarity], result of:
          0.008757716 = score(doc=2877,freq=20.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20142901 = fieldWeight in 2877, product of:
              4.472136 = tf(freq=20.0), with freq of:
                20.0 = termFreq=20.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0390625 = fieldNorm(doc=2877)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

The Information Content (IC) of a concept is a fundamental dimension in computational linguistics. It enables a better understanding of concept's semantics. In the past, several approaches to compute IC of a concept have been proposed. However, there are some limitations such as the facts of relying on corpora availability, manual tagging, or predefined ontologies and fitting non-dynamic domains in the existing methods. Wikipedia provides a very large domain-independent encyclopedic repository and semantic network for computing IC of concepts with more coverage than usual ontologies. In this paper, we propose some novel methods to IC computation of a concept to solve the shortcomings of existing approaches. The presented methods focus on the IC computation of a concept (i.e., Wikipedia category) drawn from the Wikipedia category structure. We propose several new IC-based measures to compute the semantic similarity between concepts. The evaluation, based on several widely used benchmarks and a benchmark developed in ourselves, sustains the intuitions with respect to human judgments. Overall, some methods proposed in this paper have a good human correlation and constitute some effective ways of determining IC values for concepts and semantic similarity between concepts.

Type

a

0.002189429 = product of:
  0.004378858 = sum of:
    0.004378858 = product of:
      0.008757716 = sum of:
        0.008757716 = weight(_text_:a in 2878) [ClassicSimilarity], result of:
          0.008757716 = score(doc=2878,freq=20.0), product of:
            0.043477926 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.037706986 = queryNorm
            0.20142901 = fieldWeight in 2878, product of:
              4.472136 = tf(freq=20.0), with freq of:
                20.0 = termFreq=20.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0390625 = fieldNorm(doc=2878)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

The purpose of this article is to validate, through two empirical studies, a new method for automatic evaluation of written texts, called Inbuilt Rubric, based on the Latent Semantic Analysis (LSA) technique, which constitutes an innovative and distinct turn with respect to LSA application so far. In the first empirical study, evidence of the validity of the method to identify and evaluate the conceptual axes of a text in a sample of 78 summaries by secondary school students is sought. Results show that the proposed method has a significantly higher degree of reliability than classic LSA methods of text evaluation, and displays very high sensitivity to identify which conceptual axes are included or not in each summary. A second study evaluates the method's capacity to interact and provide feedback about quality in a real online system on a sample of 924 discursive texts written by university students. Results show that students improved the quality of their written texts using this system, and also rated the experience very highly. The final conclusion is that this new method opens a very interesting way regarding the role of automatic assessors in the identification of presence/absence and quality of elaboration of relevant conceptual information in texts written by students with lower time costs than the usual LSA-based methods.

Type

a