International Journal of Human Computer Interaction and Cognitive Modelling, February 2024 Volume 1, Issue 1, ISSN/ISBN-10:1-100-22321-5 Outlier Detection in Physiological Data with Stalactite Demography Representation Fatima Isiaka Department of Computer Science, Nasarawa State University, Keffi, Nigeria Abstract In recent times, outlier detection has been used to detect points that are used to consider abnormal data attributes that don’t fit in a particular pattern. In its highly practical nature, outlier detection is used in many real-world use cases. One of the most famous cases of the use of outlier detection is in financial fraud and malicious transactions for tracking abnormal relations by intelligence agencies using intelligent analytical tools. The outlier detection algorithms are very useful for any parastatals and can help to explore common outlier detection procedures and applications to big data environments. One of its major drawbacks is that it can be intimidating and look complex for an average person. This paper discusses a straight method of detecting outliers based on the categorization of data samples from a physiological study of users on e-learning and interactive webpages. There are two thousand datasets containing both synchronise physiological response and eye movement behaviour of users to dynamic contents and the major aim is to detect outlying cases due to errors in measurement and environmental factors. The dataset was categorised into data generated from the Adult response, Student response, Younger users’ response and Children’s response. The result shows there is an outlying case in children’s response data than the others because kids are not much conversant with e-learning and hence their reaction is based on straightforward tasks and mixed visual representation of eye movement and behaviour patterns. Keywords: Outlier detection, Intelligent analytical, Detection procedures, Categorised data patterns, Physiological response, Eye movement behaviour Outlier Detection with Stalactite History Received: 2nd, April 2023 Revised: 3th, April 2023 Accepted: 6th, September 2023 Published: 8th, February 2024 History Corresponding Author Fatima ISIAKA fatima.isiaka@outlook.com Corresponding Address Department of Computer Science Keffi, Nasarawa State University, Keffi, Nigeria 1 INTRODUCTION In most companies that make use of a huge amount of data, outlier detection can help to identify suspected fraudulent cases and transactions for companies that use a credit card in daily transactions, it can be used to identify abnormal brain signals that may be or indicated in the early stage to brain cancers. In most cases, anomaly or outlier detection is more likely checking the engine light of a car and gives an alert when there is a tweak, repairs, and also maintenance that needs to be looked after [1, 2, 3, 4, 5]. International Journal of Human Computer Interaction and Cognitive Modelling, February 2024