Deep Learning and text analysis in Finance
Informations
| Course number (Stud.IP): | 39915 (lecture) + 39916 (tutorial) |
| Examination number: | 262503 |
| Module: | Master Artificial Intelligence Engineering Master Wirtschaftsinformatik Master Business Administration
|
| Hours per week per semester: | 4 (lecture+ tutorial) |
| Duration of the module: | 1 semester |
| Cycle: | Every winter term |
| Expected achievement: | Exam and project |
| ECTS: | 5 |
Recommended entry requirements
Programming skills are advantageous, but not absolutely necassary. Because of the basic knowledge from the Bachelor's degree, motivated students have all the requirements to pass this course successfully.
Possibilty to retake exam
If you have failed the exam, you can retake it corresponding to § 6 of your course and examination regulations.
- Introduction in programming with Python
- Neural networks (Forward, Recurrent and Convolutional) and their usage for
- prediction and classification of financial data
- composition of portfolio
- identification of specifics in financial data (by using Autoencoder)
- generation of artificial financial data (by using GANs)
- Copy research
- compromising of texts, wordfrequencies, topicmodelling, word vectors
- sentiment and classifying of texts
- Copy research of business reports, earning calls and financial news
Recommended Literature
- Deep Learning (2016) – Goodfellow, I., Bengio, Y., Courville, A.; MIT Press
- Machine Learning in Finance (2021) – Dixon, M.F., Halperin, I., Bilokon, P.; Springer Verlag
- Machine Learning for Text (2018) – Aggarwal, C. C., Springer Verlag
Methods of Deep Learning and text analytics were primary developed to be used in other scientific sectors, such as image recognition or for example the usage of chatbots. However more and more current practices and publications allow the conclusion to be drawn that there is a big potential of this method for the economical sector. The target is to get a basic knowledge of the function of the methods discussed in this course and to identify their possibilities of usage in the economical sector.
Teaching form
- Interactive lecture, including digital supporting documents and learning videos.
- Interactive tutorials, including data analysis made by your own.
Share this page
