Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
How News Media is Consumed by Users? Using ML/NLP Algorithms for Understanding News Articles Performance
Machine Learning and Natural Language Processing play an important role in tackling interesting challenges in the domain of news articles. The goal is to investigate and address a ML problem in the domain of news articles, such as article performance, recommendation, or other NLP-related challenges.
Keywords: Machine Learning, ML, Deep Learning, NLP, Natural Language Processing, Recommender Systems, Recommendations, News Article Performance, News Media, NZZ
Machine Learning (ML) and Natural Language Processing (NLP) play an important role in tackling interesting challenges in the domain of news articles. Some examples of those challenges are identifying the factors that make an article engaging, predicting the popularity of an article based on semantic and temporal factors, or providing users with personalized information on other articles that suit their individual interests and needs. The project is a collaboration with Neue Zürcher Zeitung (NZZ) to research, develop, and improve ML algorithms.
Machine Learning (ML) and Natural Language Processing (NLP) play an important role in tackling interesting challenges in the domain of news articles. Some examples of those challenges are identifying the factors that make an article engaging, predicting the popularity of an article based on semantic and temporal factors, or providing users with personalized information on other articles that suit their individual interests and needs. The project is a collaboration with Neue Zürcher Zeitung (NZZ) to research, develop, and improve ML algorithms.
The objective of this thesis is to investigate and address a ML problem in the domain of news articles, such as article performance (e.g. what makes a news article viral or engaging), article recommendation (e.g. explainability methods in news recommender systems), or other NLP-related challenges (e.g. topic modelling, NER). Develop ML model(s) based on word/document embeddings, view/engagement metrics and deep learning techniques to compute similarities between articles and user representations, and evaluate the model(s) performance on real-world news datasets.
The thesis will be overseen by Prof. Markus Gross and supervised by a Postdoc from the MTC (Dr. Saikishore Kalloori, Dr. Tatyana Ruzsics, or Dr. Laura Mascarell) and Dr. Cristina Kadar (NZZ).
The objective of this thesis is to investigate and address a ML problem in the domain of news articles, such as article performance (e.g. what makes a news article viral or engaging), article recommendation (e.g. explainability methods in news recommender systems), or other NLP-related challenges (e.g. topic modelling, NER). Develop ML model(s) based on word/document embeddings, view/engagement metrics and deep learning techniques to compute similarities between articles and user representations, and evaluate the model(s) performance on real-world news datasets.
The thesis will be overseen by Prof. Markus Gross and supervised by a Postdoc from the MTC (Dr. Saikishore Kalloori, Dr. Tatyana Ruzsics, or Dr. Laura Mascarell) and Dr. Cristina Kadar (NZZ).
For futher details, please contact Saikishore Kalloori (ssaikishore@ethz.ch) or Laura Mascarell (laura.mascarellespuny@inf.ethz.ch)
For futher details, please contact Saikishore Kalloori (ssaikishore@ethz.ch) or Laura Mascarell (laura.mascarellespuny@inf.ethz.ch)