Live Commenting of Football Games Using Machine Learning and Natural Language Generation
(2020) In Master's Theses in Mathematical Sciences FMAM05 20202Mathematics (Faculty of Engineering)
- Abstract
- In this degree project, automatic live comments of the events in a football game are generated in text format. The events are detected using machine learning, where CNNs are fit using audio recordings and player positions of games. Suitable features are extracted, where several models are fit to detect different types of events. The results indicate that the detections of sound powers and referee whistles are sensitive to the arena, where difficult to determine the event of the referee whistle. However, ongoing attacks are detected accurately. The detected events are commented using natural language generation, where the comments are generated using data-to-text generation. The results indicate that the complexity of the comments is... (More)
- In this degree project, automatic live comments of the events in a football game are generated in text format. The events are detected using machine learning, where CNNs are fit using audio recordings and player positions of games. Suitable features are extracted, where several models are fit to detect different types of events. The results indicate that the detections of sound powers and referee whistles are sensitive to the arena, where difficult to determine the event of the referee whistle. However, ongoing attacks are detected accurately. The detected events are commented using natural language generation, where the comments are generated using data-to-text generation. The results indicate that the complexity of the comments is sensitive to the information able to be extracted of the events. (Less)
- Popular Abstract
- Historically, live commenting of football games has required a human commentator, where mostly high profile games have been commented. However, the increase of applications for machine learning and natural language generation the past recent years, calls for automatic live commenting. In particular, for games not normally commented by human voiced commentators. This degree project investigates this problem, where machine learning is used for event detection and natural language generation for commenting the events.
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9042563
- author
- Grönvall, Marcus LU
- supervisor
- organization
- alternative title
- Live-kommentering av fotbollsmatcher med hjälp av maskininlärning och naturlig språkgenerering
- course
- FMAM05 20202
- year
- 2020
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- Live, Commenting, Football, Machine Learning, ML, Natural Language Generation, NLG, Convolutional Neural Network, CNN, Data-to-Text Generation, Spectrogram, Mel-Frequency Cepstral Coefficients, MFCC, Hard Negative Mining, Spiideo
- publication/series
- Master's Theses in Mathematical Sciences
- report number
- LUTFMA-3437-2021
- ISSN
- 1404-6342
- other publication id
- 2021:E5
- language
- English
- id
- 9042563
- date added to LUP
- 2021-05-04 13:19:49
- date last changed
- 2021-05-04 13:19:49
@misc{9042563, abstract = {{In this degree project, automatic live comments of the events in a football game are generated in text format. The events are detected using machine learning, where CNNs are fit using audio recordings and player positions of games. Suitable features are extracted, where several models are fit to detect different types of events. The results indicate that the detections of sound powers and referee whistles are sensitive to the arena, where difficult to determine the event of the referee whistle. However, ongoing attacks are detected accurately. The detected events are commented using natural language generation, where the comments are generated using data-to-text generation. The results indicate that the complexity of the comments is sensitive to the information able to be extracted of the events.}}, author = {{Grönvall, Marcus}}, issn = {{1404-6342}}, language = {{eng}}, note = {{Student Paper}}, series = {{Master's Theses in Mathematical Sciences}}, title = {{Live Commenting of Football Games Using Machine Learning and Natural Language Generation}}, year = {{2020}}, }