Modeling positive and negative feedback for improving document retrieval

Document Retrieval and Cluster Based Indexing using Rider Spider Monkey Optimization Algorithm

“…Different existing document retrieval techniques are surveyed. Hao S et al [1] introduced an expectation maximization algorithm to enhance the retrieval performance of the documents. The negative and the positive feedback were integrated by using the query language model to achieve valuable in document retrieval.…”

“…ISSN: 2277-3878, Volume-8 Issue-6, March 2020 B. Challenges  Retrieving the exactly right documents using the short query is a major challenging task in the document retrieval process, as the average person uses the short query to express and explain their expectation to retrieve the documents [1].  Reducing the usage space while retrieving the documents based on the user query is a challenging task associated with the document retrieval process.…”

“…Pseudo relevance feedback mechanism (PRF) is used to make high representation in terms of short query and also assure to enhance the retrieval performance. PRF mechanism assures that the documents retrieved in the top ranked lists are equivalent to the user query and therefore these documents are used in the form of feedback to retrieve the appropriate information [1] [9]. These methods are most effective in the document retrieval process by enhancing the searching efficiency [1].…”

“…Since, the off-topic information is extracted by using the feedback process, which further integrates the retrieved documents with the original query. The appropriate feedback for the documents are effectively selected by assigning the weights to each information document, placing the feedback terms for various documents using the weight, and to design the language model for eliminating the effect of distraction caused by the observations [1] [10] [11]. Such methods are most effective in focusing and retrieving the feedback based documents, which are may be negative or positive feedback, and hence it attained better document retrieval performance based on the query relevant to the user.…”

“…In general, the irrelevant documents attain better retrieval performance rather than the relevant documents. Since, the negative document contains the positive terms, the term distribution affects the PRF methods [1] [14] [15]. Several algorithms, like Frequent itemset based hierarchical clustering (FIHC) [18], Lazy associative tag recommender (LATR) [20], Topic document clustering (TDC) [19], Frequent itemset mining (FIM) [21] [22], and Hierarchical frequent term based clustering (HFTC) [17] are utilized to select the frequent itemset [8].…”