Conference Papers

Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506

Browse

Author: search.filters.author.Sriram, A.Subject: search.filters.subject.Explainable A

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Analyzing Banking Services Applicability Using Explainable Artificial Intelligence
    (Association for Computing Machinery, 2022) Sriram, A.; Gorti, S.S.; Amin, E.G.; Anand Kumar, M.A.
    Over the last few years, the banking sector has had a pivotal role to play in the global economy, comprising of about 24% of the global GDP and employing millions of people worldwide. Banks have a wide array of products and services to offer, ranging from ATMs, Tele-Banking, Credit Cards, Debit cards, Electronic Fund Transfers (EFT), Internet Banking, Mobile Banking, etc. Machine learning is a method of data analysis that automates analytical model building and can be an essential decision support tool for banks in providing services to certain customers and to help in improving customer satisfaction and experience based on collected data. In this study, we made use of several machine learning models and Artificial Neural Networks (ANN) to help banks make predictions about timely customer loan repayment and customer satisfaction. We explored different machine learning algorithms and have performed SHAP analysis, which has helped make conclusions about the significant features driving these decisions. © 2022 ACM.
  • No Thumbnail Available
    Item
    Exploring the Impact of External Factors on Ride-Hailing Demand: A Predictive Modelling Approach
    (The Society for the Study of Artificial Intelligence and Simulation of Behaviour, 2023) Sriram, A.; Ananthanarayana, V.S.
    This paper presents a comprehensive study on the usage of Uber in different markets, with a focus on understanding the impact of demographic factors, public transit proximity, weather and extreme events on the demand for Uber ride-hailing services. This study involves application of Explainable AI techniques for feature selection among multiple data sources to model external factors on the Uber ride usage. Furthermore, factors such as weather and local events are used for ride usage forecasting using spatiotemporal aspects and extreme event analysis. The results of this study showed that certain factors like demography, proximity of public transit play a role in shaping the usage patterns of Uber. Also, extreme events, such as weather conditions and local events, were found to have a significant impact on the demand for Uber services. This study provides valuable insights for Uber, similar ride-hailing services and policymakers for optimal resource allocation, and lays the foundation for further research on the relationship between transportation services and various contextual factors. © AISB Convention 2023.All rights reserved.