The automated categorization (or classification) of texts into predefined categories has witnessed a booming interest in the last ten years, due to the increased availability of documents in digital form and the ensuing need to organize them. In the research community the dominant approach to this problem is based on machine learning techniques: a general inductive process automatically builds a classifier by learning, from a set of preclassified documents, the characteristics of the categories. The advantages of this approach over the knowledge engineering approach (consisting in the manual definition of a classifier by domain experts) are a very good effectiveness, considerable savings in terms of expert labor power, and straightforward portability to different domains. This survey discusses the main approaches to text categorization that fall within the machine learning paradigm. We will discuss in detail issues pertaining to three different problems, namely document representation, classifier construction, and classifier evaluation.

. This paper shows that the accuracy of learned text classifiers can be improved by augmenting a small number of labeled training documents with a large pool of unlabeled documents. This is important because in many text classification problems obtaining training labels is expensive, while large quantities of unlabeled documents are readily available. We introduce an algorithm for learning from labeled and unlabeled documents based on the combination of Expectation-Maximization (EM) and a naive Bayes classifier. The algorithm first trains a classifier using the available labeled documents, and probabilistically labels the unlabeled documents. It then trains a new classifier using the labels for all the documents, and iterates to convergence. This basic EM procedure works well when the data conform to the generative assumptions of the model. However these assumptions are often violated in practice, and poor performance can result. We present two extensions to the algorithm that improve ...

Recent work in text classification has used two different first-order probabilistic models for classification, both of which make the naive Bayes assumption. Some use a multi-variate Bernoulli model, that is, a Bayesian Network with no dependencies between words and binary word features (e.g. Larkey and Croft 1996; Koller and Sahami 1997). Others use a multinomial model, that is, a uni-gram language model with integer word counts (e.g. Lewis and Gale 1994; Mitchell 1997). This paper aims to clarify the confusion by describing the differences and details of these two models, and by empirically comparing their classification performance on five text corpora. We find that the multi-variate Bernoulli performs well with small vocabulary sizes, but that the multinomial performs usually performs even better at larger vocabulary sizes---providing on average a 27% reduction in error over the multi-variate Bernoulli model at any vocabulary size.

Text categorization – the assignment of natural language texts to one or more predefined categories based on their content – is an important component in many information organization and management tasks. We compare the effectiveness of five different automatic learning algorithms for text categorization in terms of learning speed, realtime classification speed, and classification accuracy. We also examine training set size, and alternative document representations. Very accurate text classifiers can be learned automatically from training examples. Linear Support Vector Machines (SVMs) are particularly promising because they are very accurate, quick to train, and quick to evaluate. 1.1 Keywords Text categorization, classification, support vector machines, machine learning, information management.

Information filtering systems are designed for unstructured or semistructured data, as opposed to database applications, which use very structured data. The systems also deal primarily with textual information, but they may also entail images, voice, video or other data types that are part of multimedia information systems. Information filtering systems also involve a large amount of data and streams of incoming data, whether broadcast from a remote source or sent directly by other sources. Filtering is based on descriptions of individual or group information preferences, or profiles, that typically represent long-term interests. Filtering also implies removal of data from an incoming stream rather than finding data in the stream; users see only the data that is extracted. Models of information retrieval and filtering, and lessons for filtering from retrieval research are presented.

The Rocchio relevance feedback algorithm is one of the most popular and widely applied learning methods from information retrieval. Here, a probabilistic analysis of this algorithm is presented in a text categorization framework. The analysis gives theoretical insight into the heuristics used in the Rocchio algorithm, particularly the word weighting scheme and the similarity metric. It also suggests improvements which lead to a probabilistic variant of the Rocchio classifier. The Rocchio classifier, its probabilistic variant, and a naive Bayes classifier are compared on six text categorization tasks. The results show that the probabilistic algorithms are preferable to the heuristic Rocchio classifier not only because they are more well-founded, but also because they achieve better performance.

This paper examines the use of inductive learning to categorize natural language documents into predefined content categories. Categorization of text is of increasing importance in information retrieval and natural language processing systems. Previous research on automated text categorization has mixed machine learning and knowledge engineering methods, making it difficult to draw conclusions about the performance of particular methods. In this paper we present empirical results on the performance of a Bayesian classifier and a decision tree learning algorithm on two text categorization data sets. We find that both algorithms achieve reasonable performance and allow controlled tradeoffs between false positives and false negatives. The stepwise feature selection in the decision tree algorithm is particularly effective in dealing with the large feature sets common in text categorization. However, even this algorithm is aided by an initial prefiltering of features, confirming the results...

this article, we will investigate the performance of two recently implemented machine-learning algorithms on a number of large text categorization problems. The two algorithms considered are set-valued RIPPER, a recent rule-learning algorithm [Cohen A earlier version of this article appeared in Proceedings of the 19th Annual International ACM Conference on Research and Development in Information Retrieval (SIGIR) pp. 307--315

In this paper, we compare learning techniques based on statistical classification to traditional methods of relevance feedback for the document routing problem. We consider three classification techniques which have decision rules that are derived via explicit error minimization: linear discriminant analysis, logistic regression, and neural networks. We demonstrate that the classifiers perform 1015 % better than relevance feedback via Rocchio expansion for the TREC-2 and TREC-3 routing tasks. Error minimization is difficult in high-dimensional feature spaces because the convergence process is slow and the models are prone to overfitting. We use two different strategies, latent semantic indexing and optimal term selection, to reduce the number of features. Our results indicate that features based on latent semantic indexing are more effective for techniques such as linear discriminant analysis and logistic regression, which have no way to protect against overfitting. Neural networks perform equally well with either set of features and can take advantage of the additional information available when both feature sets are used as input.

. This paper shows that the accuracy of learned text classifiers can be improved by augmenting a small number of labeled training documents with a large pool of unlabeled documents. This is significant because in many important text classification problems obtaining classification labels is expensive, while large quantities of unlabeled documents are readily available. We present a theoretical argument showing that, under common assumptions, unlabeled data contain information about the target function. We then introduce an algorithm for learning from labeled and unlabeled text, based on the combination of Expectation-Maximization with a naive Bayes classifier. The algorithm first trains a classifier using the available labeled documents, and probabilistically labels the unlabeled documents. It then trains a new classifier using the labels for all the documents, and iterates. Experimental results, obtained using text from three different real-world tasks, show that the use of unlabeled...