On linear mixture of expert approaches to information retrieval

2015

Information Retrieval (IR) System is very complex in nature due to the complex interactions between documents and queries, which means that the matching of document representations and query representations is not straightforward. The Genetic Algorithm (GA) is widely used in IR systems to improve the effectiveness such systems. This study uses the Vector Space Model (VSM) and the Extended Boolean Model (EBM) to compute the similarities between queries and documents. Two fitness functions are proposed in this paper: One as fitness function and the other as adaptive mutation. Then comparing each of these functions with a number of ratio mutations that have been introduced to get better results. The experimental results reveal that the proposed cosine function outperformed other fitness models.

Information Processing & Management, 2000

This paper proposes a novel approach to automatically retrieve keywords and then uses genetic algorithms to adapt the keyword weights. One of the contributions of the paper is to combine the Pat-tree-based keyword extraction for Chinese information retrieval, ACM SIGIR'97, Philadelphia, PA, US, pp. 50±59) to retrieve keywords. The approach extracts bigrams from documents and uses the bigrams to construct a PAT-tree to retrieve keywords. The proposed approach can retrieve any type of keywords such as technical keywords and a person's name. Eectiveness of the proposed approach is demonstrated by comparing how eective are the keywords found by both this approach and the PAT-tree based approach. This comparison reveals that our keyword retrieval approach is as accurate as the PAT-tree based approach, yet our approach is faster and uses less memory. The study then applies genetic algorithms to tune the weight of retrieved keywords. Moreover, several documents obtained from web sites are tested and experimental results are compared with those of other approaches, indicating that the proposed approach is highly promising for applications. 7

Proceedings of the SIGIR, 2004

Genetic algorithm is been adopted to implement information retrieval systems by many researchers to retrieve optimal document set based on user query. However, GA is been critiqued by premature convergence due to falling into local optimal solution. This paper proposes a new hybrid crossover technique that speeds up the convergence while preserving high quality of the retrieved documents. The proposed technique is applied to HTML documents and evaluated using precision measure. The results show that this technique is efficient in balancing between fast convergence and high quality outcome.

IEEE Internet Computing, 2001

2009

With the rapid growth of the amount of data available in electronic libraries, through Internet and enterprise network mediums, advanced methods of search and information retrieval are in demand. Information retrieval systems, designed for storing, maintaining and searching large-scale sets of unstructured documents, are the subject of intensive investigation. An information retrieval system, a sophisticated application managing underlying documentary databases, is at the core of every search engine, including Internet search services. There is a clear demand for fine-tuning the performance of information retrieval systems. One step in optimizing the information retrieval experience is the deployment of Genetic Algorithms, a widely used subclass of Evolutionary Algorithms that have proved to be a successful optimization tool in many areas. In this paper, we revise and extend genetic approaches to information retrieval leverage via the optimization of search queries. As the next trend in improving search effectiveness and user-friendliness, system interaction will use fuzzy concepts in information retrieval systems. Deployment of fuzzy technology allows stating flexible, smooth and vague search criteria and retrieving a rich set of relevance ranked documents aiming to supply the inquirer with more satisfactory answers.

Genetic algorithms are usually used in information retrieval systems (IRs) to enhance the information retrieval process, and to increase the efficiency of the optimal information retrieval in order to meet the users’ needs and help them find what they want exactly among the growing numbers of available information. The improvement of adaptive genetic algorithms helps to retrieve the information needed by the user accurately, reduces the retrieved relevant files and excludes irrelevant files. In this study, the researcher explored the problems embedded in this process, attempted to find solutions such as the way of choosing mutation probability and fitness function, and chose Cranfield English Corpus test collection on mathematics. Such collection was conducted by Cyrial Cleverdon and used at the University of Cranfield in 1960 containing 1400 documents, and 225 queries for simulation purposes. The researcher also used cosine similarity and jaccards to compute similarity between the query and documents, and used two proposed adaptive fitness function, mutation operators as well as adaptive crossover. The process aimed at evaluating the effectiveness of results according to the measures of precision and recall. Finally, the study concluded that we might have several improvements when using adaptive genetic algorithms

Information Processing & Management, 1998

Genetic algorithms (GAs) search for good solutions to a problem by operations inspired from the natural selection of living beings. Among their many uses, we can count information retrieval (IR). In this field, the aim of the GA is to help an IR system to find, in a huge documents text collection, a good reply to a query expressed by the user. The analysis of phenomena seen during the implementation of a GA for IR has brought us to a new crossover operation. This article introduces this new operation and compares it with other learning methods.

2000

People use search engines to find information they desire with the aim that their information needs will be met. Information retrieval (IR) is a field that is concerned primarily with the searching and retrieving of information in the documents and also searching the search engine, online databases, and Internet. Genetic algorithms (GAs) are robust, efficient, and optimizated methods in a wide area of search problems motivated by Darwin's principles of natural selection and survival of the fittest. This paper describes information retrieval systems (IRS) components. This paper looks at how GAs can be applied in the field of IR and specifically the relevance of genetic algorithms to internet web search. Finally, from the proposals surveyed it turns out that GA is applied to diverse problem fields of internet web search.