Mining Micro Influencer from Social Media Post with NLP and Personality and Human Values Assessment
keywords MACHINE LEARNING, NATURAL LANGUAGE PROCESSING, SOFTWARE, SOFTWARE ENGINEERING
Reference persons LUCA ARDITO
External reference persons Simone Leonardi
Research Groups DAUIN - GR-16 - SOFTWARE ENGINEERING GROUP - SOFTENG
Thesis type EXPERIMENTAL / DEVELOPMENT
Description Micro-Influencers are the emerging trend in social media marketing, companies from all over the world search for these users thanks to their high ROI, engagement, and persuasion power in their communities.
During this research thesis, you will have to develop and improve a framework that will be able to detect micro-influencers on Facebook, Instagram, and Twitter. It will compute metrics about micro-influencers’ social graph, collected through a rest API, and their personality traits and basic human values with natural language processing. It is important to also consider the localization and the language used by micro-influencers.
This platform has to be a usable Python-based Machine Learning framework that interacts with social media data through rest API. The candidate will use rest API such as Tweepy. The candidate is also required to adopt machine learning algorithms, with the exploratory possibility of deep learning using popular frameworks (PyTorch, TensorFlow, Keras, etc.) and also with state-of-the-art approaches like (Transformer based Encoders, Deep Neural Network, etc.).
Required skills Conoscenze di base di Python (ML e PNL)
Conoscenza di base di REST API
Competenze nello sviluppo di software
Disponibilità a lavorare in un ambiente orientato alla ricerca
Capacità di problem-solving
Deadline 10/12/2021 PROPONI LA TUA CANDIDATURA