Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
8 results
Search Results
Item Impact analysis of online education development and implementation using machine learning model(Bentham Science Publishers, 2024) Divakarla, U.; Chandrasekaran, K.Online education is becoming increasingly necessary and in high demand as a result of the current circumstances and the enormous expansion in internet users. Various studies have been done in this area to enhance the positive benefits of offering educational courses online. One of the most crucial concerns for learning contexts like schools and universities, especially during current epidemic period, is the prediction and analysis of students' performance since it aids in the development of practical mechanisms that enhance academic achievement and prevent dropout. Most educational institutions now place a high priority on forecasting and analysing student performance. That is necessary to assist at-risk students, ensure their retention, provide top-notch learning tools and opportunities, and enhance the university's ranking and reputation. This project aims to collect information related to online education and use Machine Learning to predict students' performance. © 2024 Bentham Science Publishers. All rights reserved.Item Machine learning approach to manage adaptive push notifications for improving user experience(Association for Computing Machinery, 2020) Madhusoodanan, A.; Anand Kumar, M.; Fraser, K.; Yousuf, B.In this modern connected world mobile phone users receive a lot of notifications. Many of the notifications are useful but several cause unwanted distractions and stress. Managing notifications is a challenging task with the large influx of notifications users receive on a daily basis. This paper proposes a machine learning approach for notification management based upon the context of the user and his/her interactions with the mobile device. Since the proposed idea is to generate personalised notifications there is no ground truth data hence performance metrics such as accuracy cannot be used. The proposed solution measures the diversity score, the click through rate score and the enticement score. © 2020 ACM.Item Predictive Intelligent System Development for Disease Classification in Diagnostic Applications(Springer Science and Business Media Deutschland GmbH, 2024) Shrivathsa, T.V.; Rao, S.S.; Karanth, P.N.; Adiga, K.; Mahabala, M.; Dakappa, P.H.; Prasad, K.With ever increasing explosion in information domain and demand for highest accuracy in medical diagnosis, the existence of a reliable, accurate prediction system is the need of the hour. In this work, an effective prediction system has been developed for accurate classification of undifferentiated ailments using a unique approach. Prediction of undifferentiated diseases at an early stage always helps in better diagnosis. Illnesses like tuberculosis, non-tubercular bacterial infection, dengue fever, non-infectious diseases have regular manifestation of fever. In present work, the uniqueness lies in the use of only temperature data of the patient being referred in predicting the nature of fever, with highest degree of accuracy, instead of several self-defined parameters over limited interval of time. The system has been developed based on artificial intelligent technique, and optimization has been achieved by assessing the performance of different classifiers available. Using prediction model with classifiers, decision can take over comparative results between different classifier algorithms. A result of predictive system defines the combination of good classifier and system developed. Accuracy score and other salient parameters describe the complete picture of the system. Predictive model development in this work proved to be one of the best assistant tools to a doctor to take call over the disease crucial period. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.Item An Approach for Integrating Behavioral Analytics and Machine Learning for Enhanced Cybersecurity(Institute of Electrical and Electronics Engineers Inc., 2024) Shivappa, P.K.; Shetty D, P.Data breaches and cyber threats have evolved into increasingly complex and stealthy forms. Conventional anomaly detection based on rules is ineffective in identifying numerous contemporary attacks. Hence, User Behavior Analysis is performed on the network traffic flow data to comprehend, model, and forecast users' actions. Nevertheless, the diversity of the methods makes their understanding exceedingly complex. Therefore, domain experts use machine learning (ML) to accomplish their goals. Thus, this paper aims to suggest an innovative architecture that can detect anomalies in the network traffic flow by analyzing user behavior. The two different sets of data are used for two-class and four-class classification. Both the data are pre-processed for duplicates, missing values, and performing encoding techniques. The correlation analysis is performed to understand the user's behavior before training the ML models. The four different ML algorithms, like Logistic regression LR, KNN, DT, and RF algorithms are applied to the pre-processed datasets. The Random Forest algorithm outperforms by achieving 100% accuracy on two- and four-class classification. The described behavioral modeling approach updates cyber threat detection to match the needs of the modern, ever-changing threat landscape. © 2024 IEEE.Item Accuracy Comparison of Logistic Regression and Decision Tree Prediction Models Using Machine Learning Technique(Springer Science and Business Media Deutschland GmbH, 2025) Tantri, B.R.; Bhat, S.With the advancements in data science and machine learning, it has become beneficial for scientists, technologists, social scientists, and businessmen to adopt the latest developments in machine learning into their domains to make important decisions about their problems of interest. The biggest advantage of machine learning algorithms in such fields is their prediction capability. Statistical tools in powerful machine-learning languages like R have led to simpler solutions to more complex problems. Various models are in use in the process of making decisions and predictions. The most commonly used model in many situations is the regression model. Herein, it is intended to use the logistic regression model and the decision tree model in the prediction of binary categorical variables. R programming is used in the development of these prediction models. It is intended to compare the accuracy of the two models by using the confusion matrices. Two different datasets have been used for the prediction using these models and their comparisons. It has been observed that prediction using a decision tree model has a better accuracy as compared to that of a logistic regression model. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.Item Condition monitoring of face milling tool using K-star algorithm and histogram features of vibration signal(Elsevier B.V., 2016) Madhusudana, C.K.; Kumar, H.; Narendranath, S.This paper deals with the fault diagnosis of the face milling tool based on machine learning approach using histogram features and K-star algorithm technique. Vibration signals of the milling tool under healthy and different fault conditions are acquired during machining of steel alloy 42CrMo4. Histogram features are extracted from the acquired signals. The decision tree is used to select the salient features out of all the extracted features and these selected features are used as an input to the classifier. K-star algorithm is used as a classifier and the output of the model is utilised to study and classify the different conditions of the face milling tool. Based on the experimental results, K-star algorithm is provided a better classification accuracy in the range from 94% to 96% with histogram features and is acceptable for fault diagnosis. © 2016 Karabuk UniversityItem A machine-learning approach for classifying Indian internet shoppers(Henry Stewart Publications, 2022) Majhi, R.; Sugasi, R.P.This paper identifies the key factors that influence Indian consumers to shop online. The study uses data collected via questionnaire survey to segment consumers with shared behaviours into groups, with the results of this clustering used to train radial basis function neural networks, decision trees and random forest models. The performance of these classification models is then assessed and compared with the conventional statistical-based naïve Bayes method and logistic regression. The study finds that the random forest method provides the greatest accuracy for the segmentation of online consumers, followed by naïve Bayes and decision trees methods. The behavioural patterns identified in this study may be leveraged in real-world situations. © 2022, Henry Stewart Publications. All rights reserved.Item Decision Tree Model for Predicting Exposure Temperature and Retention Period-Dependent Behavior of Blended Concrete(Springer Science and Business Media Deutschland GmbH, 2023) Kulkarni, K.S.; Babu Narayan, K.S.; Yaragal, S.C.The major objective of the study is to estimate the behavior of blended concrete at various sustained exposure temperatures and retention times. The study examines the properties of four different types of concrete mixes, including unblended and blended mixes with fly ash and ground granulated blast furnace slag used to partially replace cement at exposure temperatures between 100 °C and 800 °C for varying exposure times of 1, 2, and 3 h. Concrete quality has been evaluated using measurements of density, porosity, and ultrasonic pulse velocity. Residual compressive and splitting tensile strengths have also been determined. The experimental study indicates that blended concrete has better fire-endurance characteristics than unblended concrete. The exposure temperature and retention time dependent behavior of unblended and blended concrete is predicted using classification and regression decision tree techniques. © 2023, The Author(s), under exclusive licence to Shiraz University.