Causality model for text data with a hierarchical topic structure
This study describes a method for constructing a causality model from text data, such as review data. Topic modeling is useful to find these evaluation factors from text data. The method based on hierarchical latent Dirichlet allocation is useful because it automatically constructs relationships amo...
Saved in:
Main Authors: | , , |
---|---|
Format: | Default Conference proceeding |
Published: |
2021
|
Subjects: | |
Online Access: | https://hdl.handle.net/2134/13203674.v1 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
rr-article-13203674 |
---|---|
record_format |
Figshare |
spelling |
rr-article-132036742021-03-23T00:00:00Z Causality model for text data with a hierarchical topic structure Takuro Ogawa (9612653) Hideyasu Shimadzu (2567326) Ryosuke Saga (9612656) Bayesian rose trees Causality analysis Topic model Hierarchical topic structure This study describes a method for constructing a causality model from text data, such as review data. Topic modeling is useful to find these evaluation factors from text data. The method based on hierarchical latent Dirichlet allocation is useful because it automatically constructs relationships among topics. However, the depth of each topic in a hierarchical structure is the same even if the contents differ for each topic. Accordingly, the method can generate less important topics that are not worth analyzing. To solve this problem, we construct a hierarchical topic structure with different depths and more important topics by using Bayesian rose trees. In the experiment, the values of the hyperparameters for constructing a hierarchical topic structure are estimated by using evaluation indexes for causal analysis. In addition, the experiment compares the proposed method with related approaches to demonstrate the usefulness of this model. 2021-03-23T00:00:00Z Text Conference contribution 2134/13203674.v1 https://figshare.com/articles/conference_contribution/Causality_model_for_text_data_with_a_hierarchical_topic_structure/13203674 All Rights Reserved |
institution |
Loughborough University |
collection |
Figshare |
topic |
Bayesian rose trees Causality analysis Topic model Hierarchical topic structure |
spellingShingle |
Bayesian rose trees Causality analysis Topic model Hierarchical topic structure Takuro Ogawa Hideyasu Shimadzu Ryosuke Saga Causality model for text data with a hierarchical topic structure |
description |
This study describes a method for constructing a causality model from text data, such as review data. Topic modeling is useful to find these evaluation factors from text data. The method based on hierarchical latent Dirichlet allocation is useful because it automatically constructs relationships among topics. However, the depth of each topic in a hierarchical structure is the same even if the contents differ for each topic. Accordingly, the method can generate less important topics that are not worth analyzing. To solve this problem, we construct a hierarchical topic structure with different depths and more important topics by using Bayesian rose trees. In the experiment, the values of the hyperparameters for constructing a hierarchical topic structure are estimated by using evaluation indexes for causal analysis. In addition, the experiment compares the proposed method with related approaches to demonstrate the usefulness of this model. |
format |
Default Conference proceeding |
author |
Takuro Ogawa Hideyasu Shimadzu Ryosuke Saga |
author_facet |
Takuro Ogawa Hideyasu Shimadzu Ryosuke Saga |
author_sort |
Takuro Ogawa (9612653) |
title |
Causality model for text data with a hierarchical topic structure |
title_short |
Causality model for text data with a hierarchical topic structure |
title_full |
Causality model for text data with a hierarchical topic structure |
title_fullStr |
Causality model for text data with a hierarchical topic structure |
title_full_unstemmed |
Causality model for text data with a hierarchical topic structure |
title_sort |
causality model for text data with a hierarchical topic structure |
publishDate |
2021 |
url |
https://hdl.handle.net/2134/13203674.v1 |
_version_ |
1797822066340134912 |