An official website of the United States government
Dynamic topic models (DTM) are commonly used for mining latent topics in evolving web corpora. In this paper, we note that a major limitation of the conventional DTM based models is that they assume a predetermined and fixed scale of topics. In reality, however, topics may have varying spans and topics of multiple scales can co-exist in a single web or social media data stream. Therefore, DTMs that assume a fixed epoch length may not be able to effectively capture latent topics and thus negatively affect accuracy. In this paper, we propose a Multi-Scale Dynamic Topic Model (MS-DTM) and a complementary Incremental Multi-Scale Dynamic Topic Model (IMS-DTM) inference method that can be used to capture latent topics and their dynamics simultaneously, at different scales. In this model, topic specific feature distributions are generated based on a multi-scale feature distribution of the previous epochs; moreover, multiple scales of the current epoch are analyzed together through a novel multi-scale incremental Gibbs sampling technique. We show that the proposed model significantly improves efficiency and effectiveness compared to the single scale dynamic DTMs and prior models that consider only multiple scales of the past.
more »
« less
DynaMiTE: Discovering Explosive Topic Evolutions with User Guidance
Dynamic topic models (DTMs) analyze text streams to capture the evolution of topics. Despite their popularity, existing DTMs are either fully supervised, requiring expensive human annotations, or fully unsupervised, generating topic evolutions that often do not cater to a user’s needs. Further, the topic evolutions produced by DTMs tend to contain generic terms that are not indicative of their designated time steps. To address these issues, we propose the task of discriminative dynamic topic discovery. This task aims to discover topic evolutions from temporal corpora that distinctly align with a set of user-provided category names and uniquely capture topics at each time step. We solve this task by developing DynaMiTE, a framework that ensembles semantic similarity, category indicative, and time indicative scores to produce informative topic evolutions. Through experiments on three diverse datasets, including the use of a newly-designed human evaluation experiment, we demonstrate that DynaMiTE is a practical and efficient framework for helping users discover high-quality topic evolutions suited to their interests.
more »
« less
- PAR ID:
- 10467022
- Publisher / Repository:
- Association for Computational Linguistics
- Date Published:
- Page Range / eLocation ID:
- 194 to 217
- Subject(s) / Keyword(s):
- Dynamic topic models, text streams, topic evolutions, discriminative dynamic topic discovery, ensembles semantic similarity, category indicative, and time indicative scores
- Format(s):
- Medium: X
- Location:
- Toronto, Canada
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Mining a set of meaningful and distinctive topics automatically from massive text corpora has broad applications. Existing topic models, however, typically work in a purely unsupervised way, which often generate topics that do not fit users’ particular needs and yield suboptimal performance on downstream tasks. We propose a new task, discriminative topic mining, which leverages a set of user-provided category names to mine discriminative topics from text corpora. This new task not only helps a user understand clearly and distinctively the topics he/she is most interested in, but also benefits directly keyword-driven classification tasks. We develop CatE, a novel category-name guided text embedding method for discriminative topic mining, which effectively leverages minimal user guidance to learn a discriminative embedding space and discover category representative terms in an iterative manner. We conduct a comprehensive set of experiments to show that CatE mines highquality set of topics guided by category names only, and benefits a variety of downstream applications including weakly-supervised classification and lexical entailment direction identification.more » « less
-
Proc. 2023 ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining (Ed.)Instead of relying on human-annotated training samples to build a classifier, weakly supervised scientific paper classification aims to classify papers only using category descriptions (e.g., category names, category-indicative keywords). Existing studies on weakly supervised paper classification are less concerned with two challenges: (1) Papers should be classified into not only coarse-grained research topics but also fine-grained themes, and potentially into multiple themes, given a large and fine-grained label space; and (2) full text should be utilized to complement the paper title and abstract for classification. Moreover, instead of viewing the entire paper as a long linear sequence, one should exploit the structural information such as citation links across papers and the hierarchy of sections and paragraphs in each paper. To tackle these challenges, in this study, we propose FuTex, a framework that uses the cross-paper network structure and the in-paper hierarchy structure to classify full-text scientific papers under weak supervision. A network-aware contrastive fine-tuning module and a hierarchyaware aggregation module are designed to leverage the two types of structural signals, respectively. Experiments on two benchmark datasets demonstrate that FuTex significantly outperforms competitive baselines and is on par with fully supervised classifiers that use 1,000 to 60,000 ground-truth training samples.more » « less
-
Proceedings of the Sixteenth (Ed.)Instead of mining coherent topics from a given text corpus in a completely unsupervised manner, seed-guided topic discovery methods leverage user-provided seed words to extract distinctive and coherent topics so that the mined topics can better cater to the user’s interest. To model the semantic correlation between words and seeds for discovering topic-indicative terms, existing seedguided approaches utilize different types of context signals, such as document-level word co-occurrences, sliding window-based local contexts, and generic linguistic knowledge brought by pre-trained language models. In this work, we analyze and show empirically that each type of context information has its value and limitation in modeling word semantics under seed guidance, but combining three types of contexts (i.e., word embeddings learned from local contexts, pre-trained language model representations obtained from general-domain training, and topic-indicative sentences retrieved based on seed information) allows them to complement each other for discovering quality topics. We propose an iterative framework, SeedTopicMine, which jointly learns from the three types of contexts and gradually fuses their context signals via an ensemble ranking process. Under various sets of seeds and on multiple datasets, SeedTopicMine consistently yields more coherent and accurate topics than existing seed-guided topic discovery approaches.more » « less
-
The objective of this paper is to propose and test a system analytics framework based on social sensing and text mining to detect topic evolution associated with the performance of infrastructure systems in disasters. Social media, like Twitter, as active channels of communication and information dissemination, provide insights into real-time information and first-hand experience from affected areas in mass emergencies. While the existing studies show the importance of social sensing in improving situational awareness and emergency response in disasters, the use of social sensing for detection and analysis of infrastructure systems and their resilience performance has been rather limited. This limitation is due to the lack of frameworks to model the events and topics (e.g., grid interruption and road closure) evolution associated with infrastructure systems (e.g., power, highway, airport, and oil) in times of disasters. The proposed framework detects infrastructure-related topics of the tweets posted in disasters and their evolutions by integrating searching relevant keywords, text lemmatization, Part-of-Speech (POS) tagging, TF-IDF vectorization, topic modeling by using Latent Dirichlet Allocation (LDA), and K-Means clustering. The application of the proposed framework was demonstrated in a study of infrastructure systems in Houston during Hurricane Harvey. In this case study, more than sixty thousand tweets were retrieved from 150-mile radius in Houston over 39 days. The analysis of topic detection and evolution from user-generated data were conducted, and the clusters of tweets pertaining to certain topics were mapped in networks over time. The results show that the proposed framework enables to summarize topics and track the movement of situations in different disaster phases. The analytics elements of the proposed framework can improve the recognition of infrastructure performance through text-based representation and provide evidence for decision-makers to take actionable measurements.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.18653/v1/2023.findings-acl.14
