To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
A Keyword Search Based Enhanced Spatially Inverted Index List For The Health Data Retrieval |
|
Author |
Sunil Kumar Reddy, Dr.P.Govindarajulu |
| Citation |
Vol. 18 No. 2 pp. 39-47 |
|
Abstract |
Retrieval of medical information from the large datasets is one of the highly demanding and crucial task in recent days. So, various research works aimed to develop an Information Retrieval (IR) system for determining the health data. Still, it remains with the major limitations of inefficient data handling, reduced accuracy, and increased searching error. Thus, this research work aims to develop a keyword based IR system by implementing an Enhanced Spatially Inverted Index List (ESIIL). In this system, the medical dataset, namely, disease symptom knowledge database is taken for analysis, which is preprocessed at the initial stage by performing the stop words removal and stemming processes. After that, the Parts of Speech (POS) tagging is used to extract the keywords from the query. Then, the probability scores, namely, Term Frequency (TF) and Inverse Document Frequency (IDF) are computed based on the similarity attributes. Here, the leaf to root based order of search is implemented, which improves the retrieval efficiency of searching. Moreover, two types of information retrieval is performed in this work such as most exactly matching documents, and partially matching documents related to the given query. The major benefits of this mechanism are reduced searching error, increased efficiency, and large dataset handling. In experiments, the performance results of the existing and proposed mechanisms are evaluated by using various measures. Also, the superiority of the proposed ESIIL is proved by comparing it with the existing techniques. |
|
Keywords |
Information Retrieval, Query Searching, Similarity Index (SI), Enhanced Spatially Inverted Index List (ESIIL), Term Frequency (TF), Inverse Document Frequency (IDF), and Parts of Speech (POS) Tagging. |
|
URL |
