Improving the Search Results Apart From Using the Boolean Operators

 Etiwel Mutero

Since the main purpose of an information retrieval system is essentially to capture wanted items and filter out the unwanted ones, it is therefore necessary for an information seeker to know how to improve the search results. In this paper, the author is going to discuss how an online information searcher can improve the quality of search results apart from using the Boolean operators.

According to Harter[1986] an online information retrieval is a computer associated hardware terminals, communication lines and links, modems and disk drives as well as software packages that carry out storage and retrieval functions on databases. It can also be defined as the process of searching remote databases for information using information and communication technologies, or the searching of computerized databases instead of printed indexes.

Apart from using Boolean operators, an online information searcher can use the SocialSimRank Bao et al[2007:501-510]. Bao et al proposed this first approach which ranks pages based on the semantic similarity between queries and pages. Another page ranking approach is called SocialPageRank which is similar to Page is designed to rank returned pages based on the observation that popular pages are usually annotated by popular tags and bookmarked by many active users.

Shepisten et al [2008:259-266] presented a strategy that clusters the entire space of tags to obtain sets of [semantically] related tags. Representing coherent topic areas, the obtained clusters are used to provide personalized item recommendations. Rather than item recommendation, the techniques presented in this paper follow personalized retrieval models applicable to web search, where links of search results are re-ranked according to the user’s preferences.

A re-ranking approach for categorization information retrieval was proposed by Bowen ZHENG et al [2009:1609-1616] to improve precision based on hierarchical feature selection method. The multiple feature selection method is to select features from texts that are organized in the form of hierarchy and we can re-rank information retrieval results with selected multiple features. In addition, they discussed how to use multiple features selected to re-rank information retrieval results. At last, an evaluation model is presented to prove this re-ranking approach.

Jingjing LIU et al [2009:1713-1723] proposed a personalized information retrieval system based on policy agents. For constructing such a system, the first created a personalized information retrieval model and then proposed a policy driven agent framework to implement this model. At the same time, to achieve a reasonable result list, they proposed a simple and effective rank algorithm for returned results.

Weijiang LI et al [2009:1201-1207] investigated a novel retrieval model that aims to concentrate more user’s attention on the parts of the text that possess a high density of relevant information, while, at the same time, to provide enough information to support their retrieval decisions. It is proposed that an automatically generated summary of each document, biased to a user’s query, can provide such a function. Based on the experimental results on TREC collections, it is showed the retrieval model based on summary is more effective than standard one based on document.

A strategy of the summary sentence selection for query-focused multi-documented summarization  through extracting keywords from relevant document set was proposed by Liang Ma et al [2008:20-23] It calculated the query related feature and the topic related feature for every word in relevant document set, then obtains the importance of the word by combining the two features. The score of candidate sentence is computed through the importance of words which they contain, and the modified MMR technology is used to adjust the score of the candidate sentence, then the candidate sentence with the highest score is selected as the summary sentence, till the length of the summary is enough.

Worasit Choochaiwattana et al [2009:215-219] examined the use of social annotations to improve the quality of web search. It involved two components. First, social annotations were used to index resources. Two annotations were used to index methods were proposed. Social annotations were used to improve search results ranking. Four annotation-based ranking methods were proposed. The result showed that using only annotation as an index of resources may be appropriate. Since social annotations could be viewed as a high level concept of the content, combining them to the content of resources could add some more important concepts to the resources. The result also suggested that both static feature and similar feature should be considered when using social annotations to re-rank search result.

Punam Bedi et al [2010:343-347] proposed semantically relevant retrieval and ranking of web resources that uses top N resources links returned from the search engine as seed, domain ontology to compute semantic relevance, and data from social Book marking System[SBS] to retrieve additional semantically relevant resources. Finally all retrieved resources are ranked according to the query relevancy using the Vector Space Model [VSM].The proposed approach presented is elucidated in three parts: a method that expands a posted query using semantic relevance by using ontology, an ontology, an algorithm to retrieve semantically relevant web resources by simulating human cognition using SBS, and a new approach to compute social semantic ranking of retrieved web resources. Thus it utilized collective advantages of Social Bookmark Tagging System and Semantic technologies.

Modern web search engines are expected to return a few most relevant results based on an often ambiguous user query and the enormous volume of web documents. Despite the simplicity and efficiency keyword queries always cannot accurately represent user’s real information need. The same keyword query may stand for different meanings for users with different domain backgrounds. Accordingly, users may differ significantly in their opinions about the relevant search results to a keyword query. A solution to this problem is to consider user-specific information during the search process, or the so called personalized search. Researchers have studied many kinds of personal data for personalized search. Other method of improving the search results includes recall, precision, proximity, field and truncation.

This paper proposed social information retrieval methods that improve the quality of search results. An experimental result shows that the proposed social information retrieval based on user interesting mining is effective.

Reference:

Bao,S et al, [2007] Optimizing Web Search Using Social Annotations, Alberta, Canada

Bowen,et al ,2009,A Re-ranking Approach for Categorization Information Retrieval Based on Multiple Feature Selection, JCIS Volume 5

Jingjing, et al, 2009, A Framework of Policy-driven Agent-based Personalized Information Retrieval System, JCIS Volume 5

Shepisten, et al,2008, Personalized Recommendation in Social Tagging Systems Using Hierarchical Clustering, New York, ACM Press

Weijiang, Tiejun 2009, Summary-based Model of Information Retrieval in Language Model Framework,  JCIS. Vol.5 [3]

Etiwel Mutero holds a Bsc Honours Degree in Records and Archives Management from the Zimbabwe Open University.Do you want assistance in writing your college or university assignment? You can contact Etiwel Mutero on 00263773614293 or etiwelm02@gmail.com