• Jeevanandhini.P
• Mrs.A. Nirmala

In this paper we examine a data mining approach for taking in probabilistic client behavior models from the database use logs. We propose a methodology for making an interpretation of database follows into portrayal appropriate for applying data mining methods. To stay away from this issue we propose novel method dependent on mix of choice tree classification algorithm and empirical time-subordinate component outline, by potential functions theory. The execution of the proposed method was experimentally evaluated on real-world data. The correlation with existing cutting edge data mining methods has affirmed remarkable execution of our method in prescient client behavior modeling and has exhibited competitive outcomes in anomaly detection.

INTERPRETABLE RECOMMENDATIONS FOR NORMAL BEHAVIOR AND DETECT FRAUD IN GRAPHS WITH TIME

"INTERPRETABLE RECOMMENDATIONS FOR NORMAL BEHAVIOR AND DETECT FRAUD IN GRAPHS WITH TIME", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 12, page no.46-57, December-2018, Available :http://www.jetir.org/papers/JETIR1812A08.pdf

"INTERPRETABLE RECOMMENDATIONS FOR NORMAL BEHAVIOR AND DETECT FRAUD IN GRAPHS WITH TIME", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 12, page no. pp46-57, December-2018, Available at : http://www.jetir.org/papers/JETIR1812A08.pdf