Anomaly Detection using Machine Learning: A Rule Based Classification and Ordered Regression Tree with Real Time Datasets
Jidiga Goverdhan Reddy1, Sammulal Porika2

1Jidiga Goverdhan Reddy*, Lecturer, Dept. of Technical Education ,Government of Telangana State and Research Scholar, JNTU University, Hyderabad, India.
2Sammulal Porika, Professor, Dept. computer engineering, JNTUH Nachupally, Jagitial, Affiliated to JNTU University, Hyderabad, India.
Manuscript received on April 20, 2020. | Revised Manuscript received on April 30, 2020. | Manuscript published on May 10, 2020. | PP: 108-112 | Volume-9 Issue-7, May 2020. | Retrieval Number: F4578049620/2020©BEIESP | DOI: 10.35940/ijitee.F4578.059720
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: What we use the protection of system data and user credentials is still very dispensable presently in factual applications frequently used by common people. Also losing their assets and confidence level due to lack of knowledge about usage of applications and failure to grab the abnormal behavior. How system data and user credentials are helpful to creating clone by others causes of showing anomalous behavior and don’t know to protect from the anomalies and how it is avoid. In this paper we are presenting short-lived discussion on anomaly detection and its nature of impact showing on original true datasets related to daily land transactions, medical and social networking. This paper shows the significant usage of machine learning approach applied in anomaly detection to know the fact anomalies in various datasets took from different sources. Here we are using an updated CART called Rule based Classification and Ordered Regression Tree (RBT-ORT). This method is new one with combination of Decision Tree; Rules of Random Tree giving a new adorned rule sets in classification and regression to ensure the improvement in results compare to other techniques. Our work carried out on three datasets, two are taken from UCI repository for machine learning and other one is real and original dataset Land sale data pertaining to land transactions noted in the year 2016-18. Finally the results of anomaly detection using Classification and Ordered Regression Tree compare with other machine learning methods such as ID3, C4.5, C-RT, PLSDA, CHAID, C4.5 Rule, I (Improved) – C4.5, K-Nearest neighbor and Neural Networks. 
Keywords: Anomaly, Anomaly detection, Classification, Regression.
Scope of the Article: Machine Learning