Sven Michalczyk

Status: Closed

Problem Description

The rapidly growing demand for data professionals has led to a perplex framework of Business Intelligence & Analytics related job offers. The market is mixing up old and new, emerging designations. For example, cannot be the role “Data Analyst” and “Business Analyst” used interchangeably in many cases? Or which competences does each role require? Is it the task of the Data Scientist to deploy the model he or she has trained?

Self-Service Business Intelligence & Analytics tools need to address diverse capabilities, the different roles require (Schuff et al., 2018; Abelló et al. 2013; Savinov, 2014). The range of capabilities is wide, going from top-down (less complex, more-rigid) to bottom-up (complex, more flexible). In order to design Self-Service Business Intelligence & Analytics (SSBI&A) tools, a common framework needs to be defined to create and map design principles to the roles.

Goal of Thesis

The main goal of the thesis is to define a framework of roles the subsequent research on SSBI&A can rely on. The framework shall be designed coming from bottom up by analyzing data related job offers.  Every role can be described by the tasks, the tools and the competencies which are required. More attributes have to be identified. Based on these attributes, the degree of “self-service” shall be analyzed and discussed afterward leading to the capabilities SSBI&A tools shall address. The potential to collaborate with the student working on the thesis “Data Science Lifecycles” is appreciated to consider SSBI&A roles in the data science lifecycle.

Work Packages

1. Motivate and identify the need by considering rigorous scientific journal publications (for e.g. De Mauro et al., 2017) as well as broader studies
2. Write a web-crawlers to extract job offers from the internet
3. Cluster crawled job offers and extract entities (apply for e.g. topic modeling, entity extraction related to the broad field of text mining and NLP)
4. Evaluate results by interviewing industry experts

Skills required

• High intrinsic motivation and good time management
• Good programming skills in a common data science language such as Python, R or others
• Interest in applying text mining algorithms and methods
• Fluent English (as the thesis has to be written in English)

Benefits

• Build up or extend your text mining skillset
• Great overview of the Data Science job market
• Possibility to publish your results
• Access to interview partners from industry

Contact

If you are interested, drop me an email with a short motivation statement, your CV and your current transcript of records. If you have questions before, do not hesitate to contact me.

sven.michalczyk@kit.edu

References

Abelló, A., Darmont, J., Etcheverry, L., Golfarelli, M., Mazón, J.-N., Naumann, F., Pedersen, T., Rizzi, S. B., Trujillo, J., Vassiliadis, P., and Vossen, G. 2013. “Fusion Cubes: Towards Self-Service Business Intelligence,” International Journal of Data Warehousing and Mining (9:2), pp. 66–88. (DOI: 10.4018/jdwm.2013040104).

De Mauro, Andrea & Greco, Marco & Grimaldi, Michele & Ritala, Paavo. (2017). Human resources for Big Data professions: A systematic classification of job roles and required skill sets. Information Processing & Management. 10.1016/j.ipm.2017.05.004.

Savinov, A. 2014. “ConceptMix: Self-Service Analytical Data Integration Based on the Concept-Oriented Model,” in Proceedings of the 3rd International Conference on Data Technologies and Applications (DATA 2014), pp. 78-84. (DOI: 10.5220/0005103700780084).

Schuff, D., Corral, K., Louis, R., and Schymik, G.2018. “Enabling self-service BI: A methodology and a case study for a model management warehouse,” Information Systems Frontiers (20:2), pp. 275–288. (DOI: 10.1007/s10796-016-9722-2).