Bachelor Thesis
Refine
Document Type
- Bachelor Thesis (3) (remove)
Language
- English (3) (remove)
Keywords
Institute
- Institute for Web Science and Technologies (3) (remove)
Social media platforms such as Twitter or Reddit allow users almost unrestricted access to publish their opinions on recent events or discuss trending topics. While the majority of users approach these platforms innocently, some groups have set their mind on spreading misinformation and influencing or manipulating public opinion. These groups disguise as native users from various countries to spread frequently manufactured articles, strong polarizing opinions in the political spectrum and possibly become providers of hate-speech or extremely political positions. This thesis aims to implement an AutoML pipeline for identifying second language speakers from English social media texts. We investigate style differences of text in different topics and across the platforms Reddit and Twitter, and analyse linguistic features. We employ feature-based models with datasets from Reddit, which include mostly English conversation from European users, and Twitter, which was newly created by collecting English tweets from selected trending topics in different countries. The pipeline classifies language family, native language and origin (Native or non-Native English speakers) of a given textual input. We evaluate the resulting classifications by comparing prediction accuracy, precision and F1 scores of our classification pipeline to traditional machine learning processes. Lastly, we compare the results from each dataset and find differences in language use for topics and platforms. We obtained high prediction accuracy for all categories on the Twitter dataset and observed high variance in features such as average text length especially for Balto-Slavic countries.
Next word prediction is the task of suggesting the most probable word a user will type next. Current approaches are based on the empirical analysis of corpora (large text files) resulting in probability distributions over the different sequences that occur in the corpus. The resulting language models are then used for predicting the most likely next word. State-of-the-art language models are based on n-grams and use smoothing algorithms like modified Kneser-Ney smoothing in order to reduce the data sparsity by adjusting the probability distribution of unseen sequences. Previous research has shown that building word pairs with different distances by inserting wildcard words into the sequences can result in better predictions by further reducing data sparsity. The aim of this thesis is to formalize this novel approach and implement it by also including modified Kneser-Ney smoothing.
Wikipedia is the biggest, free online encyclopaedia that can be expanded by any-one. For the users, who create content on a specific Wikipedia language edition, a social network exists. In this social network users are categorised into different roles. These are normal users, administrators and functional bots. Within the networks, a user can post reviews, suggestions or send simple messages to the "talk page" of another user. Each language in the Wikipedia domain has this type of social network.
In this thesis characteristics of the three different roles are analysed in order to learn how they function in one language network of Wikipedia and apply them to another Wikipedia network to identify bots. Timestamps from created posts are analysed to reveal noticeable characteristics referring to continuous messages, message rates and irregular behaviour of a user are discovered. Through this process we show that there exist differences between the roles for the mentioned characteristics.