Essays about: "xlnet"

Found 3 essays containing the word xlnet.

  1. 1. Mitigating Unintended Bias in Toxic Comment Detection using Entropy-based Attention Regularization

    University essay from KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Author : Fabio Camerota; [2023]
    Keywords : XLNet; BERT; Toxic Comment Classification; Entropy-based Attention Regularization; XLNet; BERT; Toxisk Kommentar Klassificering; Entropibaserad uppmärksamhetsreglering;

    Abstract : The proliferation of hate speech is a growing challenge for social media platforms, as toxic online comments can have dangerous consequences also in real life. There is a need for tools that can automatically and reliably detect hateful comments, and deep learning models have proven effective in solving this issue. READ MORE

  3. 2. Evaluation of Approaches for Representation and Sentiment of Customer Reviews

    University essay from KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Author : Stavros Giorgis; [2021]
    Keywords : machine learning; nlp; text analytics; sentiment analysis; transformers; tfidf; bow; fasttext; word2vec; bert; xlnet; roberta; maskininlärning; nlp; textanalys; sentimentanalys; transformatorer; tfidf; bow; fasttext; word2vec; bert; xlnet; roberta;

    Abstract : Classification of sentiment on customer reviews is a real-world application for many companies that offer text analytics and opinion extraction on customer reviews on different domains such as consumer electronics, hotels, restaurants, and car rental agencies. Natural Language Processing’s latest progress has seen the development of many new state-of-the-art approaches for representing the meaning of sentences, phrases, and words in the text using vector space models, so-called embeddings. READ MORE

  4. 3. Classifying and Comparing Latent Space Representation of Unstructured Log Data.

    University essay from KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Author : Bharat Sharma; [2021]
    Keywords : Machine learning; Natural language processing; Deep learning; Classification; Supervised learning; Transformer models; Sentence embeddings; Doc2Vec; Deep averaging networks.; Maskininlärning; naturligtspråkbehandling; djupinlärning; klassificering; övervakad inlärning; transformeringsmodeller; meningsinbäddningar; Doc2Vec; djupa linjärkombinerande nätverk.;

    Abstract : This thesis explores and compares various methods for producing vector representation of unstructured log data. Ericsson wanted to investigate machine learning methods to analyze logs produced by their systems to reduce the cost and effort required for manual log analysis. READ MORE