Hierarchical Topic Structure Research Articles

Topic Scaling: A Joint Document Scaling–Topic Model Approach to Learn Time-Specific Topics

This paper proposes a new methodology to study sequential corpora by implementing a two-stage algorithm that learns time-based topics with respect to a scale of document positions and introduces the concept of Topic Scaling, which ranks learned topics within the same document scale. The first stage ranks documents using Wordfish, a Poisson-based document-scaling method, to estimate document positions that serve, in the second stage, as a dependent variable to learn relevant topics via a supervised Latent Dirichlet Allocation. This novelty brings two innovations in text mining as it explains document positions, whose scale is a latent variable, and ranks the inferred topics on the document scale to match their occurrences within the corpus and track their evolution. Tested on the U.S. State Of The Union two-party addresses, this inductive approach reveals that each party dominates one end of the learned scale with interchangeable transitions that follow the parties’ term of office, while it shows for the corpus of German economic forecasting reports a shift in the narrative style adopted by economic institutions following the 2008 financial crisis. Besides a demonstrated high accuracy in predicting in-sample document positions from topic scores, this method unfolds further hidden topics that differentiate similar documents by increasing the number of learned topics to expand potential nested hierarchical topic structures. Compared to other popular topic models, Topic Scaling learns topics with respect to document similarities without specifying a time frequency to learn topic evolution, thus capturing broader topic patterns than dynamic topic models and yielding more interpretable outputs than a plain Latent Dirichlet Allocation.

Hierarchical lifelong topic modeling using rules extracted from network communities.

Topic models extract latent concepts from texts in the form of topics. Lifelong topic models extend topic models by learning topics continuously based on accumulated knowledge from the past which is updated continuously as new information becomes available. Hierarchical topic modeling extends topic modeling by extracting topics and organizing them into a hierarchical structure. In this study, we combine the two and introduce hierarchical lifelong topic models. Hierarchical lifelong topic models not only allow to examine the topics at different levels of granularity but also allows to continuously adjust the granularity of the topics as more information becomes available. A fundamental issue in hierarchical lifelong topic modeling is the extraction of rules that are used to preserve the hierarchical structural information among the rules and will continuously update based on new information. To address this issue, we introduce a network communities based rule mining approach for hierarchical lifelong topic models (NHLTM). The proposed approach extracts hierarchical structural information among the rules by representing textual documents as graphs and analyzing the underlying communities in the graph. Experimental results indicate improvement of the hierarchical topic structures in terms of topic coherence that increases from general to specific topics.

Hierarchical Afaan Oromoo News Text Classification

The advancement of the present day technology enables the production of huge amount of information. Retrieving useful information out of these huge collections necessitates proper organization and structuring. Automatic text classification is an inevitable solution in this regard. However, the present approach focuses on the flat classification, where each topic is treated as a separate class, which is inadequate in text classification where there are a large number of classes and a huge number of relevant features needed to distinguish between them.This paper explores the use of hierarchical structure for classifying a large, heterogeneous collection of Afaan oromoo News Text. The approach utilizes the hierarchical topic structure to decompose the classification task into a set of simpler problems, one at each node in the classification tree. An experiment had been conducted using a categorical data collected from Ethiopian News Agency (ENA) using SVM to see the performances of the hierarchical classifiers on Afaan Oromoo News Text. The findings of the experiment show the accuracy of flat classification decreases as the number of classes and documents (features) increases. Moreover, the accuracy of the flat classifier decreases at an increasing number of top feature set. The peak accuracy of the flat classifier was 68.84 % when the top 3 features were used. The findings of the experiment done using hierarchical classification show an increasing performance of the classifiers as we move down the hierarchy. The maximum accuracy achieved was 90.37% at level-3(last level) of the category tree. Moreover, the accuracy of the hierarchical classifiers increases at an increasing number of top feature set compared to the flat classifier. The peak accuracy was 89.06% using level three classifier when the top 15 features were used.Furthermore, the performance between flat classifier and hierarchical classifiers are compared using the same test data. Thus, it shows that use of the hierarchical structure during classification has resulted in a significant improvement of 29.42 % in exact match precision when compared with a flat classifier. Keywords : Automatic Text Classification DOI: 10.7176/NMMC/88-01 Publication date: February 29 th 2020

Understanding Big Data Through a Systematic Literature Review: The ITMI Model

The concept of Big Data in academic and professional literature has developed in a euphoric, chaotic, and unstructured manner. Decision-making is increasingly relying on Big Data, resorting to novel analytic methodologies that are applied in many different industries. This study aims to provide clarity over the Big Data phenomenon by means of a comprehensive and systematic literature review, able to produce a clear description of what Big Data is today, a structured classification of the various streams of current research, and a list of promising emerging trends. This study analyses a corpus of 4,327 articles through a novel combination of unsupervised algorithms that produces a hierarchical topic structure which empirically validates and enhances the “Information,” “Technology,” “Methods,” and “Impact” conceptual model of Big Data, identifying 17 fundamental topics and providing researchers and practitioners with a meaningful overview of the body of knowledge and a proposed research agenda.

Information Organization Patterns from Online Users in a Social Network

Recent years have seen the rise of user-generated contents (UGCs) in online social media. Diverse UGC sources and information overload are making it increasingly difficult to satisfy personalized information needs. To organize UGCs in a user-centered way, we should not only map them based on textual topics but also link them with users and even user communities. We propose a multi-dimensional framework to organize information by connecting UGCs, users, and user communities. First, we use a topic model to generate a topic hierarchy from UGCs. Second, an author-topic model is applied to learn user interests. Third, user communities are detected through a label propagation algorithm. Finally, a multi-dimensional information organization pattern is formulated based on similarities among the topic hierarchies of UGCs, user interests, and user communities. The results reveal that: 1) our proposed framework can organize information from multiple sources in a user-centered way; 2) hierarchical topic structures can provide comprehensive and in-depth topics for users; and, 3) user communities are efficient in helping people to connect with others who have similar interests.

Hierarchical topic modeling with automatic knowledge mining

Traditional topic modeling has been widely studied and popularly employed in expert systems and information systems. However, traditional topic models cannot discover structural relations among topics, thus losing the chance to explore the data more deeply. Hierarchical topic modeling has the capability of learning topics, as well as discovering the hierarchical topic structure from text data. But purely unsupervised models tend to generate weak topic hierarchies. To solve this problem, we propose a novel knowledge-based hierarchical topic model (KHTM), which can incorporate prior knowledge into topic hierarchy building. A key novelty of this model is that it can mine prior knowledge automatically from the topic hierarchies of multiple domains corpora. In this paper, the knowledge is represented as the word pairs which satisfy the requirement of frequent co-occurrence, and knowledge is organized in form of hierarchical structure. We also propose an iterative learning algorithm. For evaluation, we crawled two new multi-domain datasets and conducted comprehensive experiments. The experimental results show that our algorithm and model can generate more coherent topics, and more reasonable hierarchical structure.

Multimedia News Summarization in Search

It is a necessary but challenging task to relieve users from the proliferative news information and allow them to quickly and comprehensively master the information of the whats and hows that are happening in the world every day. In this article, we develop a novel approach of multimedia news summarization for searching results on the Internet, which uncovers the underlying topics among query-related news information and threads the news events within each topic to generate a query-related brief overview. First, the hierarchical latent Dirichlet allocation (hLDA) model is introduced to discover the hierarchical topic structure from query-related news documents, and a new approach based on the weighted aggregation and max pooling is proposed to identify one representative news article for each topic. One representative image is also selected to visualize each topic as a complement to the text information. Given the representative documents selected for each topic, a time-bias maximum spanning tree (MST) algorithm is proposed to thread them into a coherent and compact summary of their parent topic. Finally, we design a friendly interface to present users with the hierarchical summarization of their required news information. Extensive experiments conducted on a large-scale news dataset collected from multiple news Web sites demonstrate the encouraging performance of the proposed solution for news summarization in news retrieval.

Image Search Reranking With Hierarchical Topic Awareness.

With much attention from both academia and industrial communities, visual search reranking has recently been proposed to refine image search results obtained from text-based image search engines. Most of the traditional reranking methods cannot capture both relevance and diversity of the search results at the same time. Or they ignore the hierarchical topic structure of search result. Each topic is treated equally and independently. However, in real applications, images returned for certain queries are naturally in hierarchical organization, rather than simple parallel relation. In this paper, a new reranking method "topic-aware reranking (TARerank)" is proposed. TARerank describes the hierarchical topic structure of search results in one model, and seamlessly captures both relevance and diversity of the image search results simultaneously. Through a structured learning framework, relevance and diversity are modeled in TARerank by a set of carefully designed features, and then the model is learned from human-labeled training samples. The learned model is expected to predict reranking results with high relevance and diversity for testing queries. To verify the effectiveness of the proposed method, we collect an image search dataset and conduct comparison experiments on it. The experimental results demonstrate that the proposed TARerank outperforms the existing relevance-based and diversified reranking methods.

Topic transition detection about the East Japan great earthquake based on emerging modularity over time

Social media, in which individual users post their opinions and gradually build their consensus, is recognised as one of pervasive collaboration. Tracking topic transitions over time on a social media provides a rich insight into exploring its social context. This paper proposes a novel approach using a modularity measure which shows the quality of a division of a network into modules, for topic transition detection. In this method, first, significant topical terms are extracted from messages in social media. Next, a snapshot cooccurrence network is constructed at each time stamp. Then, hierarchical topic structures for each snapshot network are organised by a modularity measure. Words' similarities are considered by cosine similarity, and topic similarities are calculated using Jaccard coefficient. An experiment was conducted for messages related to the East Japan great earthquake in a buzz marketing site, and the effectiveness of our proposed method was shown.

HierarchicalTopics: Visually Exploring Large Text Collections Using Topic Hierarchies

Analyzing large textual collections has become increasingly challenging given the size of the data available and the rate that more data is being generated. Topic-based text summarization methods coupled with interactive visualizations have presented promising approaches to address the challenge of analyzing large text corpora. As the text corpora and vocabulary grow larger, more topics need to be generated in order to capture the meaningful latent themes and nuances in the corpora. However, it is difficult for most of current topic-based visualizations to represent large number of topics without being cluttered or illegible. To facilitate the representation and navigation of a large number of topics, we propose a visual analytics system--HierarchicalTopic (HT). HT integrates a computational algorithm, Topic Rose Tree, with an interactive visual interface. The Topic Rose Tree constructs a topic hierarchy based on a list of topics. The interactive visual interface is designed to present the topic content as well as temporal evolution of topics in a hierarchical fashion. User interactions are provided for users to make changes to the topic hierarchy based on their mental model of the topic space. To qualitatively evaluate HT, we present a case study that showcases how HierarchicalTopics aid expert users in making sense of a large number of topics and discovering interesting patterns of topic groups. We have also conducted a user study to quantitatively evaluate the effect of hierarchical topic structure. The study results reveal that the HT leads to faster identification of large number of relevant topics. We have also solicited user feedback during the experiments and incorporated some suggestions into the current version of HierarchicalTopics.

A Graph-Partitioning Framework for Aligning Hierarchical Topic Structures to Presentations

This paper studies the problem of imposing an existing hierarchical semantic structure onto a corresponding spoken document in which the structures are embedded, with the goal of indexing such documents for easier access. We propose a graph-partitioning framework to solve a semantic tree-to-string alignment problem through optimizing a normalized-cut criterion. We present models with different modeling capabilities and time complexities in this framework and provide experimental evidence of their performance. We relate graph partitioning to conventional dynamic time warping (DTW) as it applies to this problem, and show that the proposed framework can naturally include topic segmentation to accommodate cohesion constraints.

Browse by chunks

The overwhelming amount of Web videos returned from search engines makes effective browsing and search a challenging task. Rather than conventional ranked list, it becomes necessary to organize the retrieved videos in alternative ways. In this article, we explore the issue of topic mining and organizing of the retrieved web videos in semantic clusters. We present a framework for clustering-based video retrieval and build a visualization user interface. A hierarchical topic structure is exploited to encode the characteristics of the retrieved video collection and a semi-supervised hierarchical topic model is proposed to guide the topic hierarchy discovery. Carefully designed experiments on web-scale video dataset collected from video sharing websites validate the proposed method and demonstrate that clustering-based video retrieval is practical to facilitate users for effective browsing.

Multi-grain hierarchical topic extraction algorithm for text mining

Topic extraction from text corpus is the fundamental of many topic analysis tasks, such as topic trend prediction, opinion extraction. Since hierarchical structure is characteristics of topics, it is preferential for a topic extraction algorithm to output the topics description with this kind of structure. However, the hierarchical topic structure that is extracted by most of the current topic analysis algorithms cannot provide a meaningful description for all subtopics in the hierarchical tree. Here, we propose a new hierarchical topic extraction algorithm based on topic grain computation. By considering the distribution of word document frequency as a mixture Gaussian, an EM-like algorithm is employed to achieve the best number of mixture components, and the mean value of each component. Then topic grain is defined based on the mixture Gaussian parameters, and feature words are selected for the grain. A clustering algorithm is employed to the converted text set based on the feature words. After repeatedly applying the clustering algorithm to different converted text set, a multi-grain hierarchical topic structure with different subtopic feature words description is extracted. Experiments on two real world datasets which are collected from a news website show that the proposed algorithm can generate more meaningful multi-grain topic structure, by comparing with the current hierarchical topic clustering algorithms.

Hierarchical topic modeling with nested hierarchical Dirichlet process

This paper deals with the statistical modeling of latent topic hierarchies in text corpora. The height of the topic tree is assumed as fixed, while the number of topics on each level as unknown a priori and to be inferred from data. Taking a nonparametric Bayesian approach to this problem, we propose a new probabilistic generative model based on the nested hierarchical Dirichlet process (nHDP) and present a Markov chain Monte Carlo sampling algorithm for the inference of the topic tree structure as well as the word distribution of each topic and topic distribution of each document. Our theoretical analysis and experiment results show that this model can produce a more compact hierarchical topic structure and captures more fine-grained topic relationships compared to the hierarchical latent Dirichlet allocation model.

Taxonomy generation for text segments

It is crucial in many information systems to organize short text segments, such as keywords in documents and queries from users, into a well-formed taxonomy. In this article, we address the problem of taxonomy generation for diverse text segments with a general and practical approach that uses the Web as an additional knowledge source. Unlike long documents, short text segments typically do not contain enough information to extract reliable features. This work investigates the possibilities of using highly ranked search-result snippets to enrich the representation of text segments. A hierarchical clustering algorithm is then designed for creating the hierarchical topic structure of text segments. Text segments with close concepts can be grouped together in a cluster, and relevant clusters linked at the same or near levels. Different from traditional clustering algorithms, which tend to produce cluster hierarchies with a very unnatural shape, the algorithm tries to produce a more natural and comprehensive tree hierarchy. Extensive experiments were conducted on different domains of text segments, including subject terms, people names, paper titles, and natural language questions. The obtained experimental results have shown the potential of the proposed approach, which provides a basis for the in-depth analysis of text segments on a larger scale and is believed able to benefit many information systems.

Cumulative Index of CAAD: Current Status and Future Directions

This article discusses the Cumulative Index of CAAD ( CUMINCAD.SciX.net ) – a digital library set up in 1998 serving the CAAD-community as an important source of scientific information. During the first stage, the metadata of CAAD-related conference proceedings were compiled and published on-line, including all abstracts and approximately 50 % of the full-texts. In a subsequent step a Citation Index was created. Currently, theses and dissertations are being added to the library. Furthermore, a hierarchical topic structure was developed for automated classification of publications in the future, with topics being defined by keywords and characteristic papers. The next version of CUMINCAD, expected to be released later this year, will also feature a discussion forum, an event calendar, an option for commenting on and ranking publications as well as creating an on-line personal bibliographic review. CUMINCAD is a unique digital library in the field of CAAD serving a growing user-community. Younger doctoral degree students and junior researchers will benefit most from this edited, structured collection freely available via Internet.

Building topic hierarchy based on fuzzy relations

In this paper, we present a novel method for automatically building hierarchical topic structures of large text databases without any complicated linguistic analysis. Hierarchical relationship among categories from textual data can be discovered on the basis of term co-occurrence, which is described by fuzzy relations. Despite its simplicity, results of experiments on well-known document collections such as Yahoo directory data demonstrate the high quality of the resulting hierarchies.