Anurag Rana et al, International Journal of Computer Science and Mobile Computing, Vol.8 Issue.5, May- 2019, pg. 32-37 © 2019, IJCSMC All Rights Reserved 32 Available Online at www.ijcsmc.com International Journal of Computer Science and Mobile Computing A Monthly Journal of Computer Science and Information Technology ISSN 2320–088X IMPACT FACTOR: 6.199 IJCSMC, Vol. 8, Issue. 5, May 2019, pg.32 – 37 ANALYSIS OF TECHNIQUES USED TO DISCOVER PATTERNS FROM DATASET FOR DISEASE PREDICTION Mr. Anurag Rana 1 ; Mr. Ankur Sharma 2 ; Disha Pathania 3 Department of Computer Science and Engineering Arni University Kathgarh, (Indora)-176401 Kangra Himachal Pardesh Abstract: Disease detection by the use of technology becomes need of the hour. Lack of time and ignorance causes the problems to increase substantially. Historical medical records of persons can be used to analyse the patterns and discover the disease if any or the future outcomes in terms of disease to the person. This paper presents the comprehensive review of techniques under pattern mining used to discover distinct patterns from the given dataset. In addition sequential pattern mining is considered base to predict the diseases and techniques like pattern growth, incremental growth, prefix span etc. are comparatively analysed giving advantages and disadvantages of each. In other words Apriori based algorithms are analysed using proposed literature. Future enhancements are also suggested using the proposed literature. Keywords: Sequential pattern mining, pattern growth, prefix span 1. INTRODUCTION The critical approach of data mining used to discover normal and abnormal patterns from the database is sequential pattern mining. Data mining is the process of extraction of useful information from large database. The fetched information must be converted into user understandable form for future use. [1]Mining approaches used at different places vary according to size and complexity of problem in hand. Mining approaches useful for detecting patterns from the database includes web, text, sequential and temporal mining. Sequential pattern mining is the process of discovering patterns that are frequent within database. [2]The interest in pattern mining grown due to its ability to discover the hidden patterns within the database, that are useful for the users and cannot be extracted manually. Patterns category discovery is vital for successful interpretation of the disease. The sequential pattern mining finds out frequent pattern from the sequence database. [3] The well- known pattern mining methods are utilized for web-log analysis, medical record analysis and disease prediction. It identifies strong symptom/disease correlations which can be valuable information for the diagnosis and preventive medicine.