MMTI Theses and Dissertations (2017)

Permanent URI for this collection

http://hdl.handle.net/11071/5688

Browse

Browsing MMTI Theses and Dissertations (2017) by Author "Chebet, Patricia"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    A Mobile application for real time analysis of customer feedback in banks based on Naive Bayes model
    (Strathmore University, 2017) Chebet, Patricia
    Within a week of poor customer service, customers move to a competitor’s business. When this happens in large scale, the company is at risk of closing down or experiencing extreme losses. This is so because losing existing customers means decreased sales and degrading of brand reputation, as disappointed customers will spread the word, discouraging potential customers from that brand. This clearly indicates that customer satisfaction is a critical factor that dictates whether a company will be a success or not. Banks need to implement strategies that can help them predict customer issues before they escalate. These strategies should enable these banks to identify and solve any slight change in customer behavior in real-time. This study focuses on developing an application that is usable by banks to prompt customers for and analyse customer feedback to understand the customer’s opinion about their service using a mobile application. Currently, in banks, a customer leaves a note of how he or she views the banks services or give feedback using the “One click feedback” strategy that requires a customer to rank or rate the service to know their customers view about their services. In the proposed mobile application, real-time analysis of customer feedback is achieved through the utilisation of the Naive Bayes model. This model relies on the assumption that every feature, in this study each word in a review being classified, is independent of any other feature. To classify text to a polarity of either positive, negative or neutral, this model requires two data sets; a training set and a test set. The training set is used to train the model to classify texts and it had two files, one made up ¾ of 60 positive reviews and the other made up ¾ of 60 negative reviews. The test set is used to estimate the accuracy of the model and makes up ¼ of each the positive and negative data sets. The mobile application was developed using Agile Software Development Methodology and runs on Android platform.