A novel term weighting scheme for text classification: TF-MONO

The performance of text classification can be affected by the choice of appropriate term weighting scheme as well as other parameters. The terminology supervised term weighting scheme has become popular in recent years, as it may provide discriminative representation in vector space for text documents belonging to different classes. A term weighting scheme generally consists of three factors, namely term frequency factor, collection frequency factor, and length normalization factor. The researchers mostly have been focused on developing new collection frequency factors in term weighting studies. However, the term frequency factor has an important role, especially in supervised term weighting. In this study, we extensively analyzed the effects of using different term frequency factors on seven supervised term weighting schemes. While six of these supervised term weighting schemes were applied in the previous studies in the literature, we derived one of them from an existing feature selection method and it was not used as a weighting method before. This analysis is performed using SVM and Roccio classifiers on two widely known benchmark datasets with different characteristics. Experimental results showed that modification of term frequency factor in supervised term weighting schemes increased the performance of almost all weighting schemes. Also, term weighting schemes using square root function-based term frequency factor (SQRT_TF) are more successful than the ones using term frequency (TF) and logarithmic function-based term frequency (LOG_TF) factors. TF term frequency factor seems as the least effective one among three different term frequency factors according to the experimental results and statistical analysis.

Sentiment classification (SC) is an ongoing field of research, which involves computing opinions, sentiments, and the subjectivity of a text. It has recently been proven that imbalanced classification is challenging for the SC research community. Most existing studies assume that the balance between negative and positive samples may not be true in reality. This work describes a method to improve the problem of imbalanced sentiment classification using supervised term weighting schemes and shows how these weighting schemes can improve the performance of sentiment classification with imbalanced data, especially in the domain of multi-class classification. Nonetheless, to obtain the most appropriate term weighting schemes, five term weighting schemes are comparatively studied, namely tf-idf, tf-idf-icf, tf-rf, tf-igm, and sqrt_tf-igm. In addition to comparing several term weightings schemes, this work also compares four supervised machine learning algorithms to obtain an appropriate algorithm, including k-Nearest Neighbor (k-NN), Multinomial Naïve Bayes (MNB), Support Vector Machines (SVM) with linear, and SVM with RBF. After evaluating by F1, the performance of sqrt_tf-igm was superior to all other weighting schemes. Since the overall picture of sqrt_tf-igm returned better results than the tf-idf, tf-idf-icf, and tf-rf methods, with improved scores of F1 at 10.94%. Meanwhile, the result of sqrt_tf-igm was slightly better than tf-igm.

In this article, first a comprehensive study of the impact of term weighting schemes on the topic modeling performance (i.e., LDA and DMM) on Arabic long and short texts is presented. We investigate six term weighting methods including Word count method (standard topic models), TFIDF, PMI, BDC, CLPB, and CEW. Moreover, we propose a novel combination term weighting scheme, namely, CmTLB. We utilize the mTFIDF that takes into account the missing terms and the number of the documents in which the term appears when calculating the term weight. For further robust term weight, we combine mTFIDF with two weighting methods. We evaluate CmTLB against the studied weighting schemes by the quality of the learned topics (topic visualization and topic coherence), classification, and clustering tasks. We applied weighting schemes to Latent Dirichlet allocation (LDA) and Dirichlet multinomial mixture (DMM) on eight Arabic long and short document datasets, respectively. The experiment results outline that appropriate weighting schemes can effectively improve topic modeling performance on Arabic texts. More importantly, our proposed CmTLB significantly outperforms the other weighting schemes. Secondly, we investigate whether the Arabic stemming process can improve topic modeling performance. We study the three approaches of Arabic stemming including root-based, stem-based, and statistical approaches. We also train topic models with weighting schemes on documents after applying four stemmers related to different stemming approaches. The results outline that applying the stemming process not only reduces the dimensionality of term-document matrix leading to fast estimation process, but also show enhancement of topic modeling performance both on short and long Arabic documents. Moreover, Farasa stemmer achieves the highest performance in most cases, since it prevents the ambiguity that may happen because of the blind removal of the affixes such as in root-based or stem-based stemmers.

Multi-Tier Attention Network using Term-weighted Question Features for Visual Question Answering

Supervised learning methods are widely used in text sentiment classification. To acquire high classification performance, the effective and precise term weighting scheme plays a prime and necessary role for classification system. The traditional term weighting schemes often ignore the use of the available labeling information as the prior knowledge, which results the expressed relationships between features and class labels are not accurate and adequate enough. Hence, this paper proposes a term weighting scheme which makes use of class contribution of feature terms, and obtains a scheme which takes into account the class contribution, local distribution and global distribution of feature terms. The scheme expresses the relationships between features and class labels as a class contribution degree factor which is based on posterior probability, then combines it with the factors of frequency and inverse document frequency. Compared with a series of term weighting schemes, experimental results on a widely used dataset show that the proposed scheme can significantly improve the performance of sentiment classification.

The term weighting scheme, which is used to convert documents into vectors in the term spaces, is a vital step in automatic text categorization. The previous studies showed that term weighting schemes dominate the performance rather than the kernel functions of SVMs for the text categorization task. In this paper, we conducted experiments to compare various term weighting schemes with SVM on two widely-used benchmark data sets. We also presented a new term weighting scheme tf.rf for text categorization. The cross-scheme comparison was performed by using McNemar's tests. The controlled experimental results showed that the newly proposed tf.rf scheme is significantly better than other term weighting schemes. Compared with schemes related with tf factor alone, the idf factor does not improve or even decrease the term's discriminating power for text categorization. The binary and tf.chi representations significantly underperform the other term weighting schemes.

In information retrieval (IR) and related tasks, term weighting approaches typically consider the frequency of the term in the document and in the collection in order to compute a score reflecting the importance of the term for the document. In tasks characterized by the presence of training data (such as text classification) it seems logical that the term weighting function should take into account the distribution (as estimated from training data) of the term across the classes of interest. Although "supervised term weighting" approaches that use this intuition have been described before, they have failed to show consistent improvements. In this article, we analyze the possible reasons for this failure, and call consolidated assumptions into question. Following this criticism, we propose a novel supervised term weighting approach that, instead of relying on any predefined formula, learns a term weighting function optimized on the training set of interest; we dub this approach Learning to Weight (LTW). The experiments that we run on several well-known benchmarks, and using different learning methods, show that our method outperforms previous term weighting approaches in text classification.

Term weighting is an important task in many applications, such as information retrieval, extraction of significant words or automatic summarization. It translates the capacity of a term to discriminate a document within a collection, or a part of a document within a whole document. This paper deals with term weighting strategies in the context of lexical cohesion based topic segmentation. The aim is to propose a term weighting method which does not require any external information data. Weights are estimated from the content itself which is considered as a collection of mono-thematic documents. Two approaches are proposed and significant improvements are observed on a rich corpus covering various formats of Broadcast News shows from 8 French TV channels.

Neural bag-of-words models (NBOW) have achieved great success in text classification. They compute a sentence or document representation by mathematical operations such as simply adding and averaging over the word embedding of each sequence element. Thus, NBOW models have few parameters and require low computation cost. Intuitively, considering the important degree of each word and the word-order information for text classification are beneficial to obtain informative sentence or document representation. However, NBOW models hardly consider the above two factors when generating a sentence or document representation. Meanwhile, term weighting schemes assigning relatively high weight values to important words have exhibited successful performance in traditional bag-of-words models. However, it is still seldom used in neural models. In addition, n-grams capture word-order information in short context. In this paper, we propose a model called weighted word embedding model (WWEM). It is a variant of NBOW model introducing term weighting schemes and n-grams. Our model generates informative sentence or document representation considering the important degree of words and the word-order information. We compare our proposed model with other popular neural models on five datasets in text classification. The experimental results show that our proposed model exhibits comparable or even superior performance.

A study of supervised term weighting scheme for sentiment analysis

This study investigated two hypotheses concerning the use of anaphors in information retrieval. The first hypothesis, that anaphors tend to refer to integral concepts rather than to peripheral concepts, was well supported. Two samples of documents, one in psychology and the other in computer science, were examined by subject experts who judged the centrality of phrases which were referred to anaphorically. The second hypothesis, that various term weighting schemes are affected differently by anaphoric resolution, was also well supported. It was found that schemes which incorporate document length into the calculations produce much smaller increases in term weights for terms occurring in anaphoric resolutions than do those which do not consider document length. It is concluded that although anaphoric resolution has potential for better representing the “aboutness” of a document, care must be taken in choosing both the anaphoric classes to be resolved and the term weighting schemes to be used in measuring a document's topicality.

Term weighting strategy plays an essential role in the areas related to text processing such as text categorization and information retrieval. In such systems, term frequency, inverse document frequency, and document length normalization are important factors to be considered when a term weighting strategy is developed. Term length normalization is proposed to give equal opportunities to retrieve both lengthy documents and shorter ones. However, terms in very short documents that may be useless for users, especially in the scenario of Web information retrieval, could be assigned very high weights, resulting in a situation where shorter documents are ranked higher than lengthy documents that are more relevant to users information needs. In this research, a new R-tfidf term weighting strategy is proposed to alleviate the side effects of document length normalization. Experimental results demonstrate the proposed approach can to some extent improve the performance of text categorization.

Semantic text classification: A survey of past and recent advances

Illicit Web content filtering is a content-based analysis technique, applied to censor inappropriate contents on the Internet. Web content filtering can recognize undesirable contents through the application of AI techniques, linguistic analysis, or machine learning to classify Web pages into a set of predefined categories. However, the capacity to distinguish between useful and harmful Web content remains a major research challenge, which usually leads to the problem of under-blocking and over-blocking. Further, the extraction of best term representation for classifier presents a major limitation due to curse of dimensionality, where a feature can have the same term frequency (TF) in two or more categories but has different semantic meanings such as illicit pornography and sex education context also known as ambiguous issues. Besides, the high dimensionality of features on a Web page, even for moderate size, it has made the term representation value for classifier more complex, which affects the performance of classification. Thus, this research proposes a modified term weighting scheme (TWS) for narrative and discrete Web in order to increase the classification performance. Characteristics of pornography Web site were extracted and significant characteristics were identified and mapped against term weighting factors. Initial result revealed that other criteria such as rare feature have potential to be regarded as significant criteria in TWS technique to distinguish high-similarity Web content.

The construction of a text classifier usually involves (i) a phase of term selection, in which the most relevant terms for the classification task are identified, (ii) a phase of term weighting, in which document weights for the selected terms are computed, and (iii) a phase of classifier learning, in which a classifier is generated from the weighted representations of the training documents. This process involves an activity of learning, in which information on the membership of training documents in categories is used. Traditionally, learning enters only phases (i) and (iii). In this paper we propose instead that learning from training data should also affect phase (ii), i.e. that information on the membership of training documents to categories be used to determine term weights. We call this idea term weighting (STW). As an example, we propose a number of supervised variants of t f idf weighting, obtained by replacing the idf function with the function that has been used in phase (i) for term selection. We present experimental results obtained on the standard Reuters-21578 benchmark with one classifier learning method (support vector machines), three term selection functions (information gain, chi-square, and gain ratio), and both local and global term selection and weighting.