CEU Electronic Theses and Dissertations, 2022
| Author | Chen, Xibei |
|---|---|
| Title | Transactional Text Matching Using Semantic Models |
| Summary | My client is a B2B company providing IT services and consulting to external clients. One of their services is Procurement Analytics as a Service (PAaaS). This service offers data-driven insights regarding savings opportunities to external clients by focusing on the volume aggregation of certain products and services. At the end of the pipeline, they create a dashboard for their clients that they can analyze with the help of Business Intelligence officers. Their clients can then use the insights to negotiate better terms for their future procurement. My project is about developing a proof of concept. I would explore various state-of-the-art Natural Language Processing (NLP) techniques and Machine Learning algorithms to improve the existing transactional text matching model’s prediction performance. The new model ideally can more accurately predict the similarity between transactional text data and the assigned UNSPSC category. The motivation for my client to develop such proof of concept is that incorporating a better performing transactional text matching model will reduce data validation workload, and ensure more accurate categorization of purchases, thus contributing to client satisfaction. |
| Supervisor | Ariño de la Rubia, Eduardo |
| Department | Economics MSc |
| Full text | https://www.etd.ceu.edu/2022/chen_xibei.pdf |
Visit the CEU Library.
© 2007-2021, Central European University