#### DMCA

## University of Lugano at TREC 2009 Blog Track

Citations: | 1 - 0 self |

### Citations

752 |
On ordered weighted averaging aggregation operators in multicriteria decisionmaking
- Yager
- 1988
(Show Context)
Citation Context ...lem and use a voting model for tackling it. 2.1 Ordered Weighted Averaging Operators in Combining Scores The ordered weighted averaging operator, commonly called OWA operator, was introduced by Yager =-=[13]-=-. OWA provides a parametrized class of mean type aggregation operators, that can generate OR operator(Max), AND operator(Min) and any other aggregation operator between them. An OWA operator of dimens... |

175 | Clustering using a similarity measure based on shared nearest neighbours
- Jarvis, Patrick
- 1973
(Show Context)
Citation Context ...ime-window In this section we present our implementation of an incremental variant of a non-hierarchical document clustering algorithm using a similarity measure based on nearest neighbors (NN-based) =-=[5, 6, 1]-=-. We preprocess every document using the following steps: (I) HTML parsing; (II) tokenization; (III) stemming; and (IV) stopwords removal. We represent a document d using the term vector model where d... |

154 |
A computational approach to fuzzy quantifiers in natural languages
- Zadeh
- 1983
(Show Context)
Citation Context ... ) − Q( ), i = 1, 2, ..., n (4) n n where n is the number of operands to be combined, and Q is the fuzzy linguistic quantifier. We use the following definition for the Q function as suggested by Zadeh=-=[15]-=-: 8 >< 0, if r < a r − a Q(r) = >: b − a , if a ≤ r ≤ b (5) 1, if r > b with a, b, r ∈ [0, 1]. We used parameter (a, b) with three different values, (0, 0.5), (0.3, 0.8) and (0.5, 1), as three quantif... |

103 | Learning approaches for detecting and tracking news events
- Yang, Carbonell, et al.
- 1999
(Show Context)
Citation Context ...reports it in a very short time window. The second factor has the following motivation. It is known that news stories discussing the same event tend to be temporally proximate across the news streams =-=[14]-=-. Therefore we use a combine similarity measure that increases with the content similarity and decreases with the temporal distance. Let ∆t(i, j) be the temporal distance between documents di and dj, ... |

71 | Overview of the trec-2007 blog track
- Macdonald, Ounis, et al.
- 2007
(Show Context)
Citation Context ... be processed. Due to the importance of this information, TREC (Text REtrieval Conference) has started a new track for blog analysis including opinion detection, polarity mining and blog distillation =-=[7, 11]-=-. In the remainder of this paper we will explain our approach in faceted blog distillation in section 2. Our approach to top stories identification is explained in section 3. We provide conclusions in... |

67 | Finding clusters of different sizes, shapes, and densities in noisy, high dimensional data
- Ertoz, Steinbach, et al.
- 2003
(Show Context)
Citation Context ...ime-window In this section we present our implementation of an incremental variant of a non-hierarchical document clustering algorithm using a similarity measure based on nearest neighbors (NN-based) =-=[5, 6, 1]-=-. We preprocess every document using the following steps: (I) HTML parsing; (II) tokenization; (III) stemming; and (IV) stopwords removal. We represent a document d using the term vector model where d... |

56 | Overview of the TREC-2006 blog track
- Ounis, Rijke, et al.
- 2006
(Show Context)
Citation Context ... be processed. Due to the importance of this information, TREC (Text REtrieval Conference) has started a new track for blog analysis including opinion detection, polarity mining and blog distillation =-=[7, 11]-=-. In the remainder of this paper we will explain our approach in faceted blog distillation in section 2. Our approach to top stories identification is explained in section 3. We provide conclusions in... |

51 | Retrieval and feedback models for blog feed search
- Elsas, Arguello, et al.
(Show Context)
Citation Context ...thors view it as ad-hoc search and consider each blog as a long document created by concatenating all postings together. Other researchers treat it as the resource ranking problem in federated search =-=[4]-=-. They view the blog search problem as the task of ranking collec. tions of blog posts rather than single documents. A similar approach has been used in [12], where they again consider a blog as a col... |

46 | Regularizing ad hoc retrieval scores
- Diaz
(Show Context)
Citation Context ...e relationship between them. The idea behind score regularization is that in accordance with the Clustering Hypothesis, related documents should have similar scores for the same query. The authors of =-=[2, 10]-=- propose general models for smoothing document scores based on this hypothesis. In [2], Diaz models the problem in terms of optimization. The goal is to calculate for each document a new (smoothed) sc... |

27 | A general optimization framework for smoothing language models on graph structures
- Mei, Zhang, et al.
- 2008
(Show Context)
Citation Context ...e relationship between them. The idea behind score regularization is that in accordance with the Clustering Hypothesis, related documents should have similar scores for the same query. The authors of =-=[2, 10]-=- propose general models for smoothing document scores based on this hypothesis. In [2], Diaz models the problem in terms of optimization. The goal is to calculate for each document a new (smoothed) sc... |

15 | Umass at TREC 2007 blog distillation task
- Seo, Croft
- 2008
(Show Context)
Citation Context ...esource ranking problem in federated search [4]. They view the blog search problem as the task of ranking collec. tions of blog posts rather than single documents. A similar approach has been used in =-=[12]-=-, where they again consider a blog as a collection of postings and use resource selection approaches. Their intuition is that finding relevant blogs is similar to finding relevant collections in a dis... |

11 | Dynamic pattern mining: An incremental data clustering approach
- Chung, McLeod
- 2005
(Show Context)
Citation Context ...ctor associated with it. Yager introduced two measure for characterizing OWA operator[13]. The first one is called orness and is defined as: orness(W ) = 1 n − 1 i=1 nX (n − i)wi (2) i=1 orness(W ) ∈ =-=[0, 1]-=- which characterizes the degree to which the operator behaves like an or operator. The second measure is dispersion and is defined as nX dispersion(W ) = − wi ln(wi) (3) i=1and measures the degree to... |

6 |
University of Texas School of Information at TREC 2007
- Efron, Turnbull, et al.
- 2007
(Show Context)
Citation Context ...LATION Blog distillation is the problem of retrieving relevant blogs (as a collection of posts) to a given query. The blog distillation task has been approached from many different points of view. In =-=[3]-=-, the authors view it as ad-hoc search and consider each blog as a long document created by concatenating all postings together. Other researchers treat it as the resource ranking problem in federated... |