Available Topics

Information

The following topics are only sample topics that have already been completed. We are generally open to similar topics and questions. Please discuss the details with the supervisor.

Thesis advisor: Matthias Palmer, matthias.palmer@uni-goettingen.de

Main themes:

  • Data Analytics, esp. Topic Mining and Sentiment Analysis

  • Decision Support Systems, esp. in the context of capital markets


Example topic: Identification of topics in Twitter data and their relationship to capital market data: A quantitative analysis based on topic mining methods

A combination of a sentiment analysis on the basis of tweets and of an event study, e.g. with regard to company-related press releases, is able to reveal particularly noteworthy topics for a specific company context. Based on this approach, topic mining methods shall be used to illustrate how a direct identification of market-relevant topics based on Twitter data can be achieved. In addition to the choice of appropriate capital market ratios, for example abnormal returns or trading volumes, the choice of a suitable topic mining algorithm is particularly important. This algorithm should be able to deal with the text length of tweets to arrive at a meaningful representation of the tweets by means of topics. In this thesis, in addition to the consideration of established topic mining methods, for example latent Dirichlet allocation, in particular the approach “word2vec” shall be discussed and evaluated in more detail.


Literature:

  • Aiello, L. M., Petkos, G., Martin, C., Corney, D., Papadopoulos, S., Skraba, R., Göker, A., Kompatsiaris, I., and Jaimes, A. 2013. “Sensing Trending Topics in Twitter,” IEEE Transactions on Multimedia (15:6), pp. 1268-1282.

  • Alghamdi, R., and Alfalqi, K. 2015. “A Survey of Topic Modeling in Text Mining,” International Journal of Advanced Computer Science and Applications (6:1), pp. 147-153.

  • Blei, D. M., Ng, A., and Jordan, M. 2003. “Latent Dirichlet Allocation,” International Journal of Advanced Computer Science and Applications (3), pp. 147-153.

  • Han, J., Kamber, M., and Pei, J. 2011. Data Mining: Concepts and Techniques (Morgan Kaufmann Series in Data Management Systems), 3. Ed., Oxford: Elsevier.

  • Mikolov, T., Chen, K., Corrado, G., and Dean, J. 2013. “Efficient Estimation of Word Representations in Vector Space,” arXiv preprint arXiv:1301.3781.






Thesis advisor: Jan Röder, jan.roeder@uni-goettingen.de

Main themes:

  • Text Mining

  • Representation Learning


Example topic: Representation of Text for Machine Learning – A Literature Review for the Financial Domain

Due to the digitization of social and economic transactions, large amounts of data is created. Individuals and institutions are able to collect and analyze data for various purposes. Because of the volume of the data in many cases only the automated analysis of textual data is feasible. Especially the financial domain lends itself to this type of analysis since relevant texts like financial news articles, posts in social media or conference calls for quarterly earnings may contain information that can be a determinator of future market developments. Not only the machine learning algorithms themselves evolve rapidly. Also the way how texts are represented in order to capture the rich semantical and syntactical context changes. Based on your structured literature review on text representation in the financial domain you will discuss potentials and limitations while highlighting existing research gaps.


Literatur:

  • Loughran, T., and McDonald, B. 2016. “Textual Analysis in Accounting and Finance: A Survey,” Journal of Accounting Research (54:4), pp. 1187-1230.

  • Fan, W. 2015. “Tapping the power of text mining,” Communications of the ACM (49:9), pp. 76-82.

  • Tetlock, P. C., Saar-Tsechansky, M., and Mackskasssy, S. 2003. “More Than Words: Quantifying Language to Measure Firms' Fundamentals,” The Journal of Finance (63:3), pp. 1437-1467.

  • Le, Q., and T. Mikolov 2014. “Distributed Representations of Sentences and Documents,” Proceedings of the 31st International Conference on Machine Learning, pp. 1188-1196.



Thesis advisor: Albert Torno, albert.torno@uni-goettingen.de

Main theme:

  • Digital Transformation of the Finance Industry


Example topic: Digitalization and Automation of Wealth Management – Development of a Taxonomy

The Digital Transformation of services has the potential to reduce costs and to appeal to new user groups. This leads the finance industry to invest more in digitalization and automation, e.g. in the domain of wealth management. One example are so called “Robo-Advisory-Services”, which consist of IS, that guide investors through an automated investment advisory process, recommend personalized portfolio assignments, based on their individual risk-affinity, investment goals as well as capital amount and rebalance the portfolio automatically over time. This thesis aims to provide a better understanding of the possibilities and limitations of the digitalization and automation of wealth management processes. The goal is to develop a taxonomy of the domain, which provides a structured overview of dimensions and characteristics of wealth management services and its potential IS support in research and practice. Based on an analysis of the taxonomy, it should be possible to present research and application gaps within the domain of digitalized and automized wealth management.


Literature:

  • Jung, D.; Dorner, V.; Glaser, F. & Morana, S. 2018. “Robo-Advisory - Digitalization and Automation of Financial Advisory,” Business & Information Systems Engineering (60), pp. 81-86.

  • Viceira, L.; Nolan, P.; Rogers, T. & Runco, A. 2017. “Could the Big Technology Companies of Today Be the Financial Advisers of Tomorrow?,” MIT Sloan Management Review, https://sloanreview.mit.edu/article/could-the-big-technology-companies-of-today-be-the-financial-advisers-of-tomorrow/.

  • Nickerson, R. C.; Varshney, U. & Muntermann, J. 2013. “A method for taxonomy development and its application in information systems,” European Journal of Information Systems (22), pp. 336-359.

  • Eickhoff, M., Muntermann, J. & Weinrich, T. 2017. “What Do Fintechs Actually Do? A Taxonomy of Fintech Business Models,” Proceeding of the International Conference on Information Systems, Seoul, South Korea.