## Adaptive Feedback Methods in an Extended Boolean Model (2001)

Venue: | University of Southern California, Los Angeles, USA. During |

Citations: | 5 - 0 self |

### BibTeX

@INPROCEEDINGS{Choi01adaptivefeedback,

author = {Jongpill Choi and Minkoo Kim and Vijay V. Raghavan},

title = {Adaptive Feedback Methods in an Extended Boolean Model},

booktitle = {University of Southern California, Los Angeles, USA. During},

year = {2001},

pages = {42--49}

}

### OpenURL

### Abstract

Relevance feedback methods have been used in information retrieval to generate improved query formulations based on information contained in previously retrieved documents. The relevance feedback techniques have been applied to extended Boolean query formulations as well as to vector query formulations. In this paper, we propose an adaptive way to improve the retrieval performance in an extended Boolean model. We develop a neural network model in which the weights used in extended Boolean queries can be adjusted by users' relevance feedback. Experiments are performed on a TREC collection and the results show improved performance even after applying the previous feedback methods.

### Citations

3115 |
htrodzrction to modern information retrieval
- Salton, McGill
- 1983
(Show Context)
Citation Context ...the effectiveness of a retrieval operation, it is common to compute the recall-precision values. Typically, recall-precision values are computed at fixed recall levels from 0.1 to 1.0 in steps of 0.1 =-=[1, 10]-=-. Using the reformulated queries and the adjusted queries, the precision/recall values are plotted in Figure 3, 4, and 5 for the three topics, respectively. In these figures, DNF indicates the feedbac... |

2355 | Modern Information Retrieval
- Baeza-Yates, Ribeiro-Neto
- 1999
(Show Context)
Citation Context ...the effectiveness of a retrieval operation, it is common to compute the recall-precision values. Typically, recall-precision values are computed at fixed recall levels from 0.1 to 1.0 in steps of 0.1 =-=[1, 10]-=-. Using the reformulated queries and the adjusted queries, the precision/recall values are plotted in Figure 3, 4, and 5 for the three topics, respectively. In these figures, DNF indicates the feedbac... |

848 |
Relevance feedback in information retrieval
- Rocchio
- 1971
(Show Context)
Citation Context ...iments are performed on a TREC collection and the results show improved performance even after applying the previous feedback methods. 1 Introduction Many relevance feedback methods have been studied =-=[3, 7, 8, 9]. -=-The main idea of relevance feedback is to improve the retrieval performance to reformulate the queries based on users’ judgments of relevance of the retrieved documents. They are mostly based on vec... |

599 | Improving retrieval performance by relevance feedback
- Salton, Buckley
- 1990
(Show Context)
Citation Context ...iments are performed on a TREC collection and the results show improved performance even after applying the previous feedback methods. 1 Introduction Many relevance feedback methods have been studied =-=[3, 7, 8, 9]. -=-The main idea of relevance feedback is to improve the retrieval performance to reformulate the queries based on users’ judgments of relevance of the retrieved documents. They are mostly based on vec... |

457 |
An introduction to computing with neural nets
- Lippmann
- 1987
(Show Context)
Citation Context ...based extended Boolean model. In Kim and Raghavan’s approach, they nicely mapped the p-norm based extended Boolean approach into a well known neural network model, called the multi-layered perceptro=-=n [6]-=-. This approach is different from the previous neural network approaches for information retrieval [14, 16] in at least the following two aspects. In the previous approaches, they were mainly concerne... |

86 |
A critical analysis of vector space model for information retrieval
- Raghavan, Wong
- 1986
(Show Context)
Citation Context ...iments are performed on a TREC collection and the results show improved performance even after applying the previous feedback methods. 1 Introduction Many relevance feedback methods have been studied =-=[3, 7, 8, 9]. -=-The main idea of relevance feedback is to improve the retrieval performance to reformulate the queries based on users’ judgments of relevance of the retrieved documents. They are mostly based on vec... |

68 |
Information Retrieval Systems: Characteristics, Testing and Evaluation
- Lancaster
- 1968
(Show Context)
Citation Context ...ded Boolean Model We first describe the p-norm based Boolean model. Salton et al.[12] proposed an extended Boolean model in order to overcome disadvantages in the conventional Boolean retrieval model =-=[5]. Spec-=-ifically, they introduce AND/OR 1 This research was supported in part by the Korean Dept. of MIC, Grant No. AA-00-2060-00. p p p p p p ⎡q1 a1 + q2 a2 + ..... + qn a ⎤ n sim( D, QOR( p) ) = ⎢ p p... |

63 |
New experiments in relevance feedback
- Ide
- 1971
(Show Context)
Citation Context |

56 |
Fuzzy request: An approach to weighted Boolean searches
- Bookstein
- 1980
(Show Context)
Citation Context ...he retrieval methodology in vector model is not compatible with the conventional Boolean method. To apply the advantages of the vector model to the Boolean method, the Boolean model has been extended =-=[2, 12, 13]-=-, and also the relevance feedback method for the extended Boolean model has been proposed [11]. In the relevance feedback method for the extended Boolean model, an initial Boolean query is reformulate... |

44 |
Using the cosine measure in a neural network for document retrieval
- Wilkinson, Hingston
- 1991
(Show Context)
Citation Context ...nded Boolean approach into a well known neural network model, called the multi-layered perceptron [6]. This approach is different from the previous neural network approaches for information retrieval =-=[14, 16]-=- in at least the following two aspects. In the previous approaches, they were mainly concerned with term associations. In this approach, they handle relations between concepts and Boolean expressions ... |

38 |
A mathematical model of a weighted boolean retrieval system
- Waller, Kraft
- 1979
(Show Context)
Citation Context ...he retrieval methodology in vector model is not compatible with the conventional Boolean method. To apply the advantages of the vector model to the Boolean method, the Boolean model has been extended =-=[2, 12, 13]-=-, and also the relevance feedback method for the extended Boolean model has been proposed [11]. In the relevance feedback method for the extended Boolean model, an initial Boolean query is reformulate... |

21 |
Advances feedback methods in information retrieval
- Salton, Fox, et al.
- 1985
(Show Context)
Citation Context ...pply the advantages of the vector model to the Boolean method, the Boolean model has been extended [2, 12, 13], and also the relevance feedback method for the extended Boolean model has been proposed =-=[11]. -=-In the relevance feedback method for the extended Boolean model, an initial Boolean query is reformulated in a DNF (Disjunctive Normal Form) formula based on users’ relevance feedbacks. Not only ter... |

17 |
Computation of term association by a Neural Network
- Wong, Cai, et al.
- 1993
(Show Context)
Citation Context ...nded Boolean approach into a well known neural network model, called the multi-layered perceptron [6]. This approach is different from the previous neural network approaches for information retrieval =-=[14, 16]-=- in at least the following two aspects. In the previous approaches, they were mainly concerned with term associations. In this approach, they handle relations between concepts and Boolean expressions ... |

8 | Adaptive Concept-based Retrieval Using a Neural Network
- Kim, Raghavan
(Show Context)
Citation Context ...ng the weight values with the same users’ relevance feedbacks. To adjust the weight values of queries in DNF, we extend the idea of neural network leaning method that was proposed by Kim and Raghava=-=n [4], -=-and integrate it with the p-norm based extended Boolean model. In Kim and Raghavan’s approach, they nicely mapped the p-norm based extended Boolean approach into a well known neural network model, c... |