Topic Modeling with Python: What do Customers Care about Digital Banking Apps?

游騰林

游騰林

I'm Teng-Lin Yu, a data scientist from Cathay Financial Holdings. I have over eight years of hands-on experience in data analysis and machine/deep learning models, having worked in the telecommunications and financial industries. I enjoy applying data science techniques to solve real-world problems rather than theoretical research. I also share my practical experience at technology conferences in my spare time. My ideal work style is to be a consultant, helping businesses improve operational performance by sharing my data science project experience and techniques. If you are interested in data science applications, please feel free to connect with me.

    Abstract

    網路銀行App 的使用體驗是各間銀行在發展數位金融的一項成功關鍵,直接影響到客戶對於企業的忠誠度。然而要分析客戶的使用體驗並不是一件容易的事情,雖然 Google Play / Apple Store 的應用市集已經提供了 APP 的滿意度分數了,但是滿意度分數卻過度的簡化了使用體驗背後所包含的訊息。我們從滿意度分數就推敲出各 APP 獲得高/低分的原因,以及找出需要優化的方向。因此如何分析與應用非結構化的客戶留言資料就成為相當重要的課題 在這次的演講中我將以台灣 19 間銀行的 APP 在 Google Play 上共 31,474 筆評論,以真實的資料示範透過 Python 進行主題模型(Topic modeling)的方法以及過程中的成功與失敗的經驗。最終從文本中萃取出不同的討論話題,並為各個話題找出有意義的重要關鍵詞,讓公司能深入了解 APP 使用體驗與優缺點。也期望透過這次分享讓資料科學家、數據分析師在面對這類情境時能掌握透過 Python 進行 Topic Modeling 的流程以及成功關鍵

    Description

    Location

    R0

    Date

    Day 1 • 02:50-03:20 (UTC)

    Language

    Chinese talk w. Chinese slides

    Level

    Intermediate

    Category

    Machine Learning