Become part of a
Jheronimus Academy of Data Science (JADS) is offering a
PhD position Uplift Modeling
The PhD proposal Uplift modeling - Optimizing personalized offering concerns the creation of predictive models for individual clients.
In a society which is largely affected by digitalization and scaling companies, it is hard for a corporate to maintain a personal relationship with its customers and truly understand a customer’s intentions, drivers and personality. Remarkably, the amount of customer behavior data is vastly expanding, offering more and more potential to actually create a customer view. However, for most data sources, there is no direct connection between measured conversion (sales) and measured behavior in the dataset.
Currently, there is still a knowledge gap on how to infer personality traits, and consecutively buying behavior (potentially as the result of campaigning activities), from these data sources. Although efforts have been made, there is no consensus on how to comprehensively describe a person’s personality or buying interests from a customer’s behavior in other channels.
To truly understand the effects of (marketing) campaigns, marketing researchers study the concept of “lift”, i.e. the true ROI of a marketing action on a key metric (e.g. retention, customer lifetime). The resulting uplift modeling in general has already made his way to the Silicon Valley, but it still lacks traction in the more traditional industries (including Telco) worldwide. Properly modeling uplift would directly benefit KPNs understanding of the effectiveness of various campaigns on retention and life time value.
In the project we focus on Bayesian deep learning and additive regression tree models to estimate the causal effects of intest (uplift).
The goal of this PhD project is to bring KPN into this new era and provide a platform that will allow its data scientists to develop cross channel and cross product customer profiles, to optimize the lift of KPN’s targeting activities. Specifically, the PhD project will provide answers to the following key questions:
How to develop a customer profile containing personality traits on both an individual as well as household level, based on cross channel behavior?
Which customer to target (household / person level) with personalized experiences and activities? (i.e. the ones with the highest lift)?
How to target this household (i.e. the offer with the highest lift)?
When to target (i.e. at the moment that yields the highest lift)?
KPN is one of the forefront runners in following the privacy-by-design paradigm. However, there are a number of use cases where the KPN private-by-design processes can be improved in a data-driven, machine-supported, and experiment-based fashion. Currently, lack of proper anonymization prevents KPN from executing these use cases.
The research will be conducted under supervision of prof.dr. M.C. Kaptein PDEng, promotor, prof.dr. M. Pechenizkiy, promoter.
The successful candidate is expected to:
- Perform scientific research in the domain described;
- Present results at (international) conferences;
- Publish results in scientific journals;
- Participate in activities of the group, mainly in 's-Hertogenbosch but sometimes at KPN the commercial partner.
- have a MSc. in Mathematics, Statistics, Computer Science, Econometrics, AI or a related discipline;
- have a strong interest in Data Engineering and Artificial Intelligence;
- have excellent analytical skills and is highly motivated and rigorous;
- have good technical understanding of models used in data engineering/science;
- have knowledge of, or a willingness to familiarize themselves with, current research into new and innovative data science techniques;
- be a fast learner, autonomous and creative, show dedication and be hard working;
- possess good communication skills and be an efficient team worker;
- be fluent in English, both spoken and written.
Conditions of employmentThe PhD student will be appointed at JADS via an employment at Eindhoven University of Technology (TU/e)
- An exciting job in a dynamic work environment
- A full-time employment for four years, with an intermediate evaluation after one year.
- To support you during your PhD and to prepare you for the rest of your career, you will have free access to a personal development program for PhD students (PROOF program).
- A gross monthly salary and benefits in accordance with the Collective Labor Agreement for Dutch Universities.
- Additionally, an annual holiday allowance of 8% of the yearly salary, plus a year-end allowance of 8.3% of the annual salary.
- The opportunity to perform cutting edge research in a large-scale joint data science project involving TiU, TU/e, JADS and a commercial partner and bringing together expertise of several senior researchers;
- A broad package of fringe benefits (including excellent technical infrastructure, savings schemes and excellent sport facilities).
Information and applicationThe Jheronimus Academy of Data Science (JADS) constitutes a unique concept in which an integrated approach to Data Science is created by combining the exact sciences of the Eindhoven University of Technology, with the social sciences of Tilburg University. JADS boasts three campuses at Tilburg, Eindhoven and Den Bosch. JADS Campus Den Bosch revolves around research, education and valorisation on data entrepreneurship.
More information about JADS can be found at www.jads.nl.
Information and application
Do you recognize yourself in this profile and would you like to know more? Please contact prof. dr. M.C. Kaptein PDEng (E: M.C.Kaptein[at]tilburguniversity.edu), or prof.dr. M. Pechenizkiy (E: m.pechenizkiy[at]tue.nl).
For information about terms of employment, please contact HR services, email@example.com or +31 40 247 3699
Please visit http://www.tue.nl/jobs to find out more about working at TU/e!
We invite you to submit an application which should include a:
You can apply by pressing the "apply now" button.
Please don't send us your application by e-mail.