Hierarchical Topic Mining Via Joint Spherical Tree And Text Embedding

Figure 1 From Hierarchical Topic Mining Via Joint Spherical Tree And We develop a novel joint tree and text embedding method along with a principled optimization procedure that allows simultaneous modeling of the category tree structure and the corpus generative process in the spherical space for effective category representative term discovery. Given a text corpus and a tree structured hierarchy described by category names, hierarchical topic mining aims to retrieve a set of terms that provide a clear description of each category.

Table 2 From Hierarchical Topic Mining Via Joint Spherical Tree And (2) we develop a joint embedding framework for hierarchical topic mining by simultaneously modeling the user provided category tree structure and the text generation process. Our comprehensive experiments show that our model, named josh, mines a high quality set of hierarchical topics with high efficiency and benefits weakly supervised hierarchical text classification tasks. The source code used for hierarchical topic mining via joint spherical tree and text embedding, published in kdd 2020. the code structure (especially file reading and saving functions) is adapted from the word2vec implementation. We develop a novel joint tree and text embedding method along with a principled optimization procedure that allows simultaneous modeling of the category tree structure and the corpus.

Figure 4 From Hierarchical Topic Mining Via Joint Spherical Tree And The source code used for hierarchical topic mining via joint spherical tree and text embedding, published in kdd 2020. the code structure (especially file reading and saving functions) is adapted from the word2vec implementation. We develop a novel joint tree and text embedding method along with a principled optimization procedure that allows simultaneous modeling of the category tree structure and the corpus. We develop a joint embedding framework for hierarchical topic mining by simultaneously modeling the user provided category tree structure and the text generation process. This work proposes a new task, hierarchical topic mining, which takes a category tree described by category names only, and aims to mine a set of representative terms for each category from a text corpus to help a user comprehend his her interested topics. We develop a novel joint tree and text embedding method along with a principled optimization procedure that allows simultaneous modeling of the category tree structure and the corpus generative process in the spherical space for effective category representative term discovery. The source code used for hierarchical topic mining via joint spherical tree and text embedding, published in kdd 2020. the code structure (especially file reading and saving functions) is adapted from the word2vec implementation.