Using Advance Natural Language Processing (NLP) on Text in Customer Feedback and Complaints to Identify Emerging Issues and Trends
Engineering, IT, Mathematics and Statistics
- This internship is able to cover project costs for domestic students only.
- The Industry Partner has implemented appropriate preparations to comply with Federal and State Government requirements regarding COVID-19 safety. Due to remote arrangements, this internship is now accepting applications from eligible PhD students nationwide.
ABOUT THE INDUSTRY PARTNER
Core to The Westpac Group’s vision is a desire to be leading service organisation. The organisation recognises that customer feedback and complaints are a key element to helping realise this vision, with complaints considered a second chance for the bank with a customer.
Over the last 12 months, The Westpac Group have invested heavily in improving the management of customer feedback and complaints, including campaigns to increase the logging of complaints from customers and bankers, investing in systems to better capture feedback, skilling up the teams that manage customer issues and investing in the use of data and analytics to help provide insight into customer issues and pain points.
At present the classification of complaints & feedback remains at high level categories, with the task of drilling down into smaller groups of complaints to understand more granular issues performed by humans reading individual items. When complaint volumes were low this was a manageable task, however with the changes outlined there has been a material rise in logged volumes making this task difficult.
WHAT’S IN IT FOR YOU?
Collaborating with a dedicated Supervisor and IT professionals at Westpac, the Intern will gain valuable commercial exposure and have valuable mentoring on the project that will improve their soft skills and improve future work readiness. Understanding the daily challenges in a Banking environment and working on deep data sets will challenge and enrich the intern’s knowledge at project completion.
RESEARCH TO BE CONDUCTED
- Analysis of text in complaints data sets and existing categorisation to understand the business problem & determine best techniques to help solve.
- Design an experiment to test the feasibility of different unsupervised NLP techniques to solve the business problem post investigation.
- Execute experiment identified in phase one
- Perform analytics using NLP techniques to help detect emerging issues and patterns within the text for each complaint category
- Work with Westpac SMEs to help ensure there is a path to productionise and regularly update outputs of models.
- Explore ways to visualise text/data (word trees, word clouds, t-SNE plots) of emerging issues found in part 2 into existing BI toolkit (Tableau). This is aimed at helping business users see patterns in the words to pinpoint and interpret.
SKILLS WISH LIST
If you’re a PhD student and meet some or all the below we want to hear from you. We strongly encourage women, indigenous and disadvantaged candidates to apply:
- Technically advanced in the use analytical programming, with a preference for python and its packages associated with Natural language Processing (NLP).
- Sound understand of the well-researched and emerging concepts in unsupervised NLP and evidence of application to solve complex problems.
- An ability to explain the concepts of NLP and its outputs to technical and non-technical stakeholder groups in both formal and informal situations.
- Background in data visualisation techniques, with any experience in visualising text viewed favourably.
As previously described, as complaint volumes grow it will become more difficult to find emerging issues and trends, particularly in the tail of distribution of complaints via traditional means. This project will demonstrate that machine learning can help to accelerate the detection of emerging pattern within unstructured text that can be visualised and passed to non-technical business users. This interpretation will help in the identification of business improvement opportunities to help contribute to Westpac desire to become a leading service business.
The intern will receive $3,000 per month of the internship, usually in the form of stipend payments.
It is expected that the intern will primarily undertake this research project during regular business hours and maintain contact with their academic mentor throughout the internship either through face-to-face or phone meetings as appropriate.
The intern and their academic mentor will have the opportunity to negotiate the project’s scope, milestones and timeline during the project planning stage.
CONNECT WITH APR.INTERN