Gaussian Mixture Variational Autoencoder for Semi-Supervised Topic Modeling
نویسندگان
چکیده
منابع مشابه
Variational Autoencoder for Semi-Supervised Text Classification
Although semi-supervised variational autoencoder (SemiVAE) works in image classification task, it fails in text classification task if using vanilla LSTM as its decoder. From a perspective of reinforcement learning, it is verified that the decoder’s capability to distinguish between different categorical labels is essential. Therefore, Semi-supervised Sequential Variational Autoencoder (SSVAE) ...
متن کاملSemi-Supervised Recursive Autoencoder
In this project, we implement the semi-supervised Recursive Autoencoders (RAE), and achieve the result comparable with result in [1] on the Movie Review Polarity dataset1. We achieve 76.08% accuracy, which is slightly lower than [1] ’s result 76.8%, with less vector length. Experiments show that the model can learn sentiment and build reasonable structure from sentence.We find longer word vecto...
متن کاملPrediction-Constrained Training for Semi-Supervised Mixture and Topic Models
Supervisory signals have the potential to make low-dimensional data representations, like those learned by mixture and topic models, more interpretable and useful. We propose a framework for training latent variable models that explicitly balances two goals: recovery of faithful generative explanations of high-dimensional data, and accurate prediction of associated semantic labels. Existing app...
متن کاملWeak Supervision for Semi-supervised Topic Modeling via Word Embeddings
Semi-supervised algorithms have been shown to improve the results of topic modeling when applied to unstructured text corpora. However, sufficient supervision is not always available. This paper proposes a new process, Weak+, suitable for use in semi-supervised topic modeling via matrix factorization, when limited supervision is available. This process uses word embeddings to provide additional...
متن کاملImproving Twitter Sentiment Analysis with Topic-Based Mixture Modeling and Semi-Supervised Training
In this paper, we present multiple approaches to improve sentiment analysis on Twitter data. We first establish a state-of-the-art baseline with a rich feature set. Then we build a topic-based sentiment mixture model with topic-specific data in a semi-supervised training framework. The topic information is generated through topic modeling based on an efficient implementation of Latent Dirichlet...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Access
سال: 2020
ISSN: 2169-3536
DOI: 10.1109/access.2020.3001184