Abstract

Saudi Arabia is fastest developing nation enjoying stability and high per capita income, thus highly influenced by urbanization inviting huge investments from international brands especially in food and clothing sector. Changes in life style have made Saudi society more prone towards disease like diabetes that is costing about 40% cases of the total population (1). Higher diabetic cases have alarmed health care organizations (both government and private) in finding the exact number of diabetic cases in extremely timely manner. Creating a unified system among hospitals, laboratories and other health care organizations is time consuming and expensive inviting researchers to look for other options. Due to improvement in community awareness among stakeholders (patients, care taker, health researcher and others), has provided an opportunity to get a real time estimate about total number of patients and getting to know about patient problems etc. This study tends to create a diabetic prediction system that will gather information from multiple sources (news, health care records, social media, news feeds, search trends and tweets) in multiple languages (Arabic , English and French) to answer two questions (1) Can online search trends and tweets be related to exact number of diabetes patients (2) Can we extract common or new symptoms for diabetes cases from these trends (3) providing a predictive picture to health care professionals and managers for creating in-time policies to avoid epidemic.(4) Finding relationship in between diabetes related search terms and diabetic cases. This study reveals that real data figures are 85% correlated to search trend thus providing a cogent proof that both internet usage and real data figures can be related. It was also observed that search trends commonly symbolize common symptoms or disease name. A cyber diabetic community can be created that can be targeted by government agencies or health organization as to create awareness about diabetes. While usage of system by community will also help in better diagnosis from search trends and hospital information.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call