Best Treatment Identification for Disease Using Machine Learning Approach in Relation to Short Text
Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2014, Vol 16, Issue 3
Abstract
Abstract: The goal of Machine Learning is to construct a computer system that can adapt and learn from their experience. Machine Learning approach helps to integrate the computer based system into the healthcare field in order to obtain best and accurate results for the system. Here the system deals with automatic identification of informative sentences from medical published by medical journals. Our main aim is to integrate machine learning in medical field and build an application that is capable of automatically identifying and disseminating disease and treatment related information, further it also identifies semantic relations that exists between diseases and treatments. In the proposed work user will search for the disease summary (disease and treatment related information) by giving symptoms as a query in the search engine. Initially when a pdf is downloaded and saved in the system it first performs per processing on the data in the document and the extracted relevant data is stored in the database. The symptoms entered by the user are further classified using SVM classifier to make the further process easier to find the semantic keyword which helps to identify the disease easily and quickly. Then the semantic keyword found is matched with the stored medical input database to identify the exact disease related to that keyword present. Once the disease related to the symptom is identified, it is sent to medical database to extract the articles pertaining to that disease. The preprocessing process involves tokenization, removal of stop words and stemming. Followed by that, relevant information is extracted using the keyword searching algorithm. The combination of BOW, NLP and biomedical concepts are put together toe identifying semantic relations that exist between diseases and treatments in biomedical sentences. Till now the best result obtain is 98.51% F-measure by OanaFrunza, for the extraction of cure and prevents relations. In our implementation of our proposed system we have used SVM classifier which gives us an improved result. The problem statement of the existing system was, it didn’t identify the best disease treatment. So the proposed solution used data mining concepts using voting algorithm to resolve the problem and find the best treatment for disease out of the treatment identified by the system.
Authors and Affiliations
Bharti E. Nerkar, , Sanjay S. Gharde
Keywords
Related Articles
- EP ID EP105162
- DOI 10.9790/0661-16370512
- Views 105
- Downloads 0
Automatic Identification of moving Objects in a Scene using Corners features
Abstract: This Paper presents an automated method to identify the moving objects. The proposed method is a combination of Corner Detectors, Area Based Tracking, K – Means clustering. The Corner Detectors used in this are...
Real-time Moving Object Detection using SURF
Abstract: Tracking and traffic monitoring are main application of moving Object detection in video. This paperpresents SURF (Speed-Up Robust Features) algorithm and real-time detection of objects using frame difference.T...
Frequent Itemset-based Text Clustering Approach to Cluster Ranked Documents
Abstract: In most of the search engines documents are retrieved and ranked on the basis of relevance. They are not necessarily ranked on the basis of similarity between the query and the respective document. The ra...
Various Methods for Object Tracking- A Review
Abstract: The object tracking is the technique which is used to track object from the image or from the video. The video consists of multiple frames and in each frame location of that object had been predicted. To predic...
Using Ensemble Methods for Improving Classification of the KDD CUP ’99 Data Set
Abstract: The KDD CUP ’99 data set has been widely used for intrusion detection and pattern mining in the last decade or so. Umpteen number of experiments pertaining to classification have been conducted on it.Many resea...