Other affiliations: Chinese Ministry of Education
Bio: Jihong Ouyang is an academic researcher from Jilin University. The author has contributed to research in topics: Topic model & Computer science. The author has an hindex of 11, co-authored 34 publications receiving 275 citations. Previous affiliations of Jihong Ouyang include Chinese Ministry of Education.
TL;DR: A common semantics topic model (CSTM) is proposed, which is a new type of topic, namely common topic, to gather the noise words in short texts to solve the sparsity and noise problems.
TL;DR: Experimental results show that weighting words can effectively improve the topic modeling performance over both short texts and normal long texts, and the proposed CEW significantly outperforms the existing term weighting schemes, since it further considers which words are informative.
Abstract: Topic models often produce unexplainable topics that are filled with noisy words. The reason is that words in topic modeling have equal weights. High frequency words dominate the top topic word lists, but most of them are meaningless words, e.g., domain-specific stopwords. To address this issue, in this paper we aim to investigate how to weight words, and then develop a straightforward but effective term weighting scheme, namely entropy weighting (EW). The proposed EW scheme is based on conditional entropy measured by word co-occurrences. Compared with existing term weighting schemes, the highlight of EW is that it can automatically reward informative words. For more robust word weight, we further suggest a combination form of EW (CEW) with two existing weighting schemes. Basically, our CEW assigns meaningless words lower weights and informative words higher weights, leading to more coherent topics during topic modeling inference. We apply CEW to Dirichlet multinomial mixture and latent Dirichlet allocation, and evaluate it by topic quality, document clustering and classification tasks on 8 real world data sets. Experimental results show that weighting words can effectively improve the topic modeling performance over both short texts and normal long texts. More importantly, the proposed CEW significantly outperforms the existing term weighting schemes, since it further considers which words are informative.
TL;DR: In LTM, the observable short texts are snippets of normal long texts generated by a given standard topic model, but their original document memberships are unknown, and with Gibbs sampling, LTM drives an adaptive aggregation process of short texts, and simultaneously estimates other latent variables of interest.
Abstract: Topic modeling for short texts faces a tough challenge, owing to the sparsity problem. An effective solution is to aggregate short texts into long pseudo-documents before training a standard topic model. The main concern of this solution is the way of aggregating short texts. A recent developed self-aggregation-based topic model (SATM) can adaptively aggregate short texts without using heuristic information. However, the model definition of SATM is a bit rigid, and more importantly, it tends to overfitting and time-consuming for large-scale corpora. To improve SATM, we propose a generalized topic model for short texts, namely latent topic model (LTM). In LTM, we assume that the observable short texts are snippets of normal long texts (namely original documents) generated by a given standard topic model, but their original document memberships are unknown. With Gibbs sampling, LTM drives an adaptive aggregation process of short texts, and simultaneously estimates other latent variables of interest. Additionally, we propose a mini-batch scheme for fast inference. Experimental results indicate that LTM is competitive with the state-of-the-art baseline models on short text topic modeling.
••17 Oct 2018
TL;DR: This paper proposes a novel Laplacian seed word topic model (LapSWTM), which significantly outperforms the existing dataless text classification algorithms and is even competitive with supervised algorithms to some extent.
Abstract: Recently, dataless text classification has attracted increasing attention. It trains a classifier using seed words of categories, rather than labeled documents that are expensive to obtain. However, a small set of seed words may provide very limited and noisy supervision information, because many documents contain no seed words or only irrelevant seed words. In this paper, we address these issues using document manifold, assuming that neighboring documents tend to be assigned to a same category label. Following this idea, we propose a novel Laplacian seed word topic model (LapSWTM). In LapSWTM, we model each document as a mixture of hidden category topics, each of which corresponds to a distinctive category. Also, we assume that neighboring documents tend to have similar category topic distributions. This is achieved by incorporating a manifold regularizer into the log-likelihood function of the model, and then maximizing this regularized objective. Experimental results show that our LapSWTM significantly outperforms the existing dataless text classification algorithms and is even competitive with supervised algorithms to some extent. More importantly, it performs extremely well when the seed words are scarce.
TL;DR: Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis.
Abstract: Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).
01 Jan 2006
TL;DR: Probability distributions of linear models for regression and classification are given in this article, along with a discussion of combining models and combining models in the context of machine learning and classification.
Abstract: Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.
01 Jan 1983
TL;DR: The results proved that the proposed improved krill herd algorithm with hybrid function achieved almost all the best results for all datasets in comparison with the other comparative algorithms.
Abstract: In this paper, a novel text clustering method, improved krill herd algorithm with a hybrid function, called MMKHA, is proposed as an efficient clustering way to obtain promising and precise results in this domain. Krill herd is a new swarm-based optimization algorithm that imitates the behavior of a group of live krill. The potential of this algorithm is high because it performs better than other optimization methods; it balances the process of exploration and exploitation by complementing the strength of local nearby searching and global wide-range searching. Text clustering is the process of grouping significant amounts of text documents into coherent clusters in which documents in the same cluster are relevant. For the purpose of the experiments, six versions are thoroughly investigated to determine the best version for solving the text clustering. Eight benchmark text datasets are used for the evaluation process available at the Laboratory of Computational Intelligence (LABIC). Seven evaluation measures are utilized to validate the proposed algorithms, namely, ASDC, accuracy, precision, recall, F-measure, purity, and entropy. The proposed algorithms are compared with the other successful algorithms published in the literature. The results proved that the proposed improved krill herd algorithm with hybrid function achieved almost all the best results for all datasets in comparison with the other comparative algorithms.
TL;DR: A taxonomy for text classification according to the text involved and the models used for feature extraction and classification is created, dealing with both the technical developments and benchmark datasets that support tests of predictions.
Abstract: Text classification is the most fundamental and essential task in natural language processing. The last decade has seen a surge of research in this area due to the unprecedented success of deep learning. Numerous methods, datasets, and evaluation metrics have been proposed in the literature, raising the need for a comprehensive and updated survey. This paper fills the gap by reviewing the state of the art approaches from 1961 to 2020, focusing on models from shallow to deep learning. We create a taxonomy for text classification according to the text involved and the models used for feature extraction and classification. We then discuss each of these categories in detail, dealing with both the technical developments and benchmark datasets that support tests of predictions. A comprehensive comparison between different techniques, as well as identifying the pros and cons of various evaluation metrics are also provided in this survey. Finally, we conclude by summarizing key implications, future research directions, and the challenges facing the research area.