Search (284 results, page 15 of 15)

7.4368593E-4 = product of:
  0.0044621155 = sum of:
    0.0044621155 = weight(_text_:in in 5209) [ClassicSimilarity], result of:
      0.0044621155 = score(doc=5209,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.07514416 = fieldWeight in 5209, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=5209)
  0.16666667 = coord(1/6)

Abstract

Chen et alia report on the design of FEATURES, a web search engine with adaptive features based on minimal relevance feedback. Rather than developing user profiles from previous searcher activity either at the server or client location, or updating indexes after search completion, FEATURES allows for index and user characterization files to be updated during query modification on retrieval from a general purpose search engine. Indexing terms relevant to a query are defined as the union of all terms assigned to documents retrieved by the initial search run and are used to build a vector space model on this retrieved set. The top ten weighted terms are presented to the user for a relevant non-relevant choice which is used to modify the term weights. Documents are chosen if their summed term weights are greater than some threshold. A user evaluation of the top ten ranked documents as non-relevant will decrease these term weights and a positive judgement will increase them. A new ordering of the retrieved set will generate new display lists of terms and documents. Precision is improved in a test on Alta Vista searches.

7.4368593E-4 = product of:
  0.0044621155 = sum of:
    0.0044621155 = weight(_text_:in in 3472) [ClassicSimilarity], result of:
      0.0044621155 = score(doc=3472,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.07514416 = fieldWeight in 3472, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3472)
  0.16666667 = coord(1/6)

Abstract

The authors address the problem of unsupervised ensemble ranking. Traditional approaches either combine multiple ranking criteria into a unified representation to obtain an overall ranking score or to utilize certain rank fusion or aggregation techniques to combine the ranking results. Beyond the aforementioned combine-then-rank and rank-then-combine approaches, the authors propose a novel rank-learn-combine ranking framework, called Interactive Ranking (iRANK), which allows two base rankers to teach each other before combination during the ranking process by providing their own ranking results as feedback to the others to boost the ranking performance. This mutual ranking refinement process continues until the two base rankers cannot learn from each other any more. The overall performance is improved by the enhancement of the base rankers through the mutual learning mechanism. The authors further design two ranking refinement strategies to efficiently and effectively use the feedback based on reasonable assumptions and rational analysis. Although iRANK is applicable to many applications, as a case study, they apply this framework to the sentence ranking problem in query-focused summarization and evaluate its effectiveness on the DUC 2005 and 2006 data sets. The results are encouraging with consistent and promising improvements.

7.4368593E-4 = product of:
  0.0044621155 = sum of:
    0.0044621155 = weight(_text_:in in 4746) [ClassicSimilarity], result of:
      0.0044621155 = score(doc=4746,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.07514416 = fieldWeight in 4746, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=4746)
  0.16666667 = coord(1/6)

Abstract

Traditional ranking models for information retrieval lack the ability to make a clear distinction between relevant and nonrelevant documents at top ranks if both have similar bag-of-words representations with regard to a user query. We aim to go beyond the bag-of-words approach to document ranking in a new perspective, by representing each document as a sequence of sentences. We begin with an assumption that relevant documents are distinguishable from nonrelevant ones by sequential patterns of relevance degrees of sentences to a query. We introduce the notion of relevance flow, which refers to a stream of sentence-query relevance within a document. We then present a framework to learn a function for ranking documents effectively based on various features extracted from their relevance flows and leverage the output to enhance existing retrieval models. We validate the effectiveness of our approach by performing a number of retrieval experiments on three standard test collections, each comprising a different type of document: news articles, medical references, and blog posts. Experimental results demonstrate that the proposed approach can improve the retrieval performance at the top ranks significantly as compared with the state-of-the-art retrieval models regardless of document type.

7.4368593E-4 = product of:
  0.0044621155 = sum of:
    0.0044621155 = weight(_text_:in in 645) [ClassicSimilarity], result of:
      0.0044621155 = score(doc=645,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.07514416 = fieldWeight in 645, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=645)
  0.16666667 = coord(1/6)

Abstract

LEarning TO Rank (LETOR) algorithms are usually trained on annotated corpora where a single relevance label is assigned to each available document-topic pair. Within the Cranfield framework, relevance labels result from merging either multiple expertly curated or crowdsourced human assessments. In this paper, we explore how to train LETOR models with relevance judgments distributions (either real or synthetically generated) assigned to document-topic pairs instead of single-valued relevance labels. We propose five new probabilistic loss functions to deal with the higher expressive power provided by relevance judgments distributions and show how they can be applied both to neural and gradient boosting machine (GBM) architectures. Moreover, we show how training a LETOR model on a sampled version of the relevance judgments from certain probability distributions can improve its performance when relying either on traditional or probabilistic loss functions. Finally, we validate our hypothesis on real-world crowdsourced relevance judgments distributions. Overall, we observe that relying on relevance judgments distributions to train different LETOR models can boost their performance and even outperform strong baselines such as LambdaMART on several test collections.