Ontology based Machine Learning using Data Mining techniques
Manish kumar Khare · Shodhgangotri · 2013
Learning, like intelligence, covers such a broad range of processes that it is difficult to define precisely. With respect to machines, it can be broadly said that a machine learns whenever it changes its structure, program, or data (based on its inputs or in response to external information) in such a manner that it‟s expected future performance improves. The objective of this proposal is to build an e-healthcare framework that is discovered by converging the semantic web and web mining techniques. This e-healthcare framework would empower the patient and physician both for taking better decision for managing health. Healthcare institutions need to target the patients with proper portfolios of services. There is a wealth of data available within the healthcare systems but they lack effective analysis tools to discover hidden relationships and trends in data. By applying web mining techniques on ontology based structure of web, useful patterns and knowledge can be discovered for making intelligent decisions for the care of individuals. Keyword: Ontology, Semantic web, Web mining, E-Healthcare, Association, Classification Introduction The healthcare environment is generally „information rich‟ but „knowledge poor‟. There is a wealth of data available within the healthcare systems but they lack effective analysis tools to discover hidden relationships and trends in data. A major challenge facing healthcare organizations (hospitals, medical centers) is the provision of quality services at affordable costs. Quality service implies diagnosing patients correctly and administering treatments that are effective. Poor clinical decisions can lead to disastrous consequences. Thus healthcare institutions need to target the patients with proper portfolios of services. Valuable knowledge can be discovered with the application of mining technique to equip patients with knowledge for managing health for disease specific better care and understanding through E-Healthcare. E-Healthcare is conceptually connecting the patients and doctors, medical information services and providing a platform to facilitate better understanding and care. The distinction between e-Healthcare and traditional healthcare is that the later can be termed as a doctor centered approach where the doctor gives consultation after analyzing the available information, past history and constraints of the patient based on his knowledge and experience. Hence doctor is the prime source of advice to the patient regarding his health problem, future care and medical prescription whereas “eHealth is a consumer-centered model of health care where stakeholders collaborate utilizing information and communication technologies (ICTs) including Internet technologies to manage health, arrange, deliver, and account for care, and manage the health care system”. Thus e-Healthcare facilitates Patient education for better health anytime and anywhere with the flexibility of time and place. Motivation & Significance of the Problem With respect to healthcare industry, there is a wealth of hidden information available that is largely untapped. The maximum percentage of data on the web are so much unstructured, heterogeneous, distributed and time variant that they can only be understood by humans, however the amount of data is gigantic that it can only be processed efficiently by machines. Therefore the great success of the current form of WWW leads to new challenges. Data Interpretation using Machines: A huge amount of data available online is interpretable by humans only, machine support in terms of interpretation of data available online is very much limited. Intelligent Data Processing: Since the size of the data in web is enormous it can only be processed by machine efficiently. However efficiency does not guarantee the intelligence; the machine support in terms of processing the data intelligently for determining useful patterns and knowledge is limited. Personalization of the Information: It is very much likely that people differ in the contents and presentations they prefer while interacting with the Web, more precisely Web organizers must know what the patients do and want. Furthermore it can be categorized in keeping track of individual preferences and facilitating recommendations as well. The motivation behind this proposal is to address these challenges in the E-Healthcare sector to turn patient specific data into knowledge and determine useful pattern that can enable health practitioners and patients to make intelligent decisions.