📌 Key facts
- When: Start anytime. Applications are open!
- How to apply: Send us an e-mail (at the end of this page) with your documents.
💡 Background
This IDP is concerned with the question how scientists are represented in German media. To answer this question, we develop the GETT (Gender Equality Tech Tool). The GETT aims to continuously monitor the presence of scientists in media, quantify their appearance and analyze how they are presented.
🦾Who We Are
yathos is a software development and consulting company with a focus on tailor made software for research and businesses. We aim to provide reliant and low maintenance software products. This ensures the future success of our customers. We provide the full service from consulting, project management, implementation, and operation of software.
🎯 Goals
Help to develop the GETT to investigate how scientists are quantitatively and qualitatively represented in German media.
Create an automated analysis process leveraging AI LLMs suited to extract relevant information from articles. (e.g. Llama2, Falcon, …).
Compare the quality of available LLMs results on the defined process.
Evaluate automated data crawling of media outlets using LLMs vs classical web scraping methods.
The aims of this project are:
- Implementing analysis tools to classify the content of the articles (text and images) regarding topics, mentioned and quoted personas using ML/AI/LLM and conventional text processing
- Implementing interfaces to participating/ cooperating data sources and/or crawlers for outlets without a dedicated interface to gather the outlets’ articles
🎓 Profile
- Skills interacting with:
- HTTP interfaces
- RESTful Services
- XML/JSON content
- SQL Databases
- Skills using/ applying:
- AI Language Models
- Classification algorithms (Nearest Neighbor, Naïve Bayes, SVM, …)
- Sentiment analysis
- Python coding experience
- Bonus: experience with Java/JavaEE, Docker
📄 Deliverables
The students must provide the developed source code, enabling GETT to use the result and alter the code if needed.
A documentation of the code must be present inline. A separate short documentation of the developed functionality is to be created.
📝 How to Apply
If you are interested, please contact Nadja Born (nadja.born@tum.de) by submitting the following documents in one PDF:
- Grade report
- Short overview of your experience in software development, including a list of coding languages and technologies in which you already have knowledge and if you have any experience in working with AI