Path Expression-based Smoothing of Query Likelihood Model for XML Element Retrieval

In this paper, we discuss the properties of statistical behavior and entropies of three smoothing methods; two well-known and one proposed smoothing method will be used on three language models in Mandarin data sets. Because of the problem of data sparseness, smoothing methods are employed to estimate the probability for each event (including all the seen and unseen events) in a language model. A set of properties used to analyze the statistical behaviors of three smoothing methods are proposed. Our proposed smoothing methods comply with all the properties. We implement three language models in Mandarin data sets and then discuss the entropy. In general, the entropies of proposed smoothing method for three models are lower than that of other two methods.

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.

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.

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.

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.

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.

BackgroundOrientation field (OF) plays a very significant role in automatic fingerprint recognition systems. Many algorithms have been proposed for the estimation of fingerprints’ OF but it is hard to solve the dilemma of correcting spurious ridge structure and avoiding singularity location deviation, especially for poor images. So far, the following drawbacks still need to be solved for OF construction methods for practical application: (1) How to adaptively choose block scales to resolve the contradiction between accuracy and anti-noise, since small scale is beneficial to accuracy but is sensitive to noise, while large scale is more resistant to noise, but the accuracy is deteriorated. (2) How to construct the genuine OF in the areas close-by singular points and to evade singularity location deviation? Current block based methods give spurious OF estimates in the area near singular points because these areas have large curvature thus the detected singular points deviate from the genuine localizations. When these singular points are used as the anchor for referencing minutiae, it makes the average error of matching or recognition even larger. Therefore, it is essentials to construct the genuine OF in the areas close-by singular points and to evade singularity deviation.MethodsTo overcome the above-mentioned limitations, a novel method, combining a weighted multi-scale composite window (WMCM) with a hierarchical smoothing strategy has been proposed for the computation of fingerprint OF. This method mainly contains two procedures: the approximate OF estimation and the hierarchical OF smoothing. In the first procedure, a series of OFs are established under multiple scales of composite windows by using a gradient based method then a coarse OF is estimated using the weight of each scale determined by a squared gradient consistency. In the second procedure, the OF is first quantized into a two-digitized orientation zone and a two-orientation-zone filtering strategy is adapted to the OF blocks based on a filtering mask obtained after eliminating the isolated blocks. In the end a similar three-digitized orientation zone is performed to obtain an accurate and smooth OF. To validate the performance, the proposed method has been applied to OF computation using the FVC2004 databases and three experiments are designed. Experiment 1 aims to validate whether the weighted multi-scale composite window can balance the dilemma of accuracy and robustness more effectively than the previous works do. Experiment 2 is designed to examine whether the hierarchical smoothing method can correct the spurious ridge flow and preserve the genuine localization of singular points. The purpose of experiment 3 is to test the performance of the proposed method on OF reconstruction in low quality fingerprint images. The fingerprint databases FVC 2004 DB1–DB4 are employed in this study.ResultsThe results of experiment I shows that the proposed method is capable to extract the information of OF reliably and it is more robust against singularity localization deviation in comparison with the other three gradient based methods. The results of experiment II indicates that the proposed smoothing method can balance the contradiction in correcting spurious ridge structures and preserving genuine singularity localization. The results of experiment III illustrates that our approach combing WMCW with the hierarchical smoothing method is capable to extract the information of OF ridge reliably and it is more robust against singularity deviation in comparison with the other three gradient based methods. In a word, the experiment results demonstrate that the proposed method can correct spurious ridge structure and meanwhile avoid singularity deviation compared with the previous works.ConclusionsA novel gradient based algorithm has been proposed which is more reliable for the estimation of the ridge information for fingerprint OF and is more accurate in preserving the singularity localization. Compared with the previously proposed gradient based methods, the advantages of the proposed RBSF lie in three aspects. Firstly a weighted multi-scale composite window is put forward to replace the single window used by conventional gradient based methods and to adaptively choose the scales of the blocks. Secondly, a hierarchical smoothing strategy is proposed to enhance the OF by using the two-orientation-zone filtering and the three-orientation-zone filtering, aiming to correct the spurious ridges and preserving the genuine location of singular points. Finally, three experiments are designed to test the proposed algorithm together with other popular gradient based methods on real fingerprint images, which are selected from different categories and all are suffering from obvious noise effects. All the experiment results show that the proposed method is superior with respect to reliable OF construction and avoiding singularity localization deviation.

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

Data sparseness has been an inherited issue of statistical language models and smoothing method is usually used to resolve the zero count problems. In this paper, we studied empirically and analyzed the well-known smoothing methods of Good-Turing and advanced Good-Turing for language models on large sizes Chinese corpus. In the paper, ten models are generated sequentially on various size of corpus, from 30 M to 300 M Chinese words of CGW corpus. In our experiments, the smoothing methods; Good-Turing and Advanced Good-Turing smoothing are evaluated on inside testing and outside testing. Based on experiments results, we analyzed further the trends of perplexity of smoothing methods, which are useful for employing the effective smoothing methods to alleviate the issue of data sparseness on various sizes of language models. Finally, some helpful observations are described in detail.

Real-time local smoothing for five-axis linear toolpath considering smoothing error constraints

Japanese large-vocabulary continuous-speech recognition using a newspaper corpus and broadcast news

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.

In this paper we present a new language model based on an odds formula, which explicitly incorporates document length as a parameter. Furthermore, a new smoothing method called exponential smoothing is introduced, which can be combined with most language models. We present experimental results for various language models and smoothing methods on a collection with large document length variation, and show that our new methods compare favorably with the best approaches known so far.

Language modeling plays a critical role for automatic speech recognition. Typically, the n-gram language models suffer from the lack of a good representation of historical words and an inability to estimate unseen parameters due to insufficient training data. In this study, we explore the application of latent semantic information (LSI) to language modeling and parameter smoothing. Our approach adopts latent semantic analysis to transform all words and documents into a common semantic space. The word-to-word, word-to-document and document-to-document relations are, accordingly, exploited for language modeling and smoothing. For language modeling, we present a new representation of historical words based on retrieval of the most relevant document. We also develop a novel parameter smoothing method, where the language models of seen and unseen words are estimated by interpolating the k nearest seen words in the training corpus. The interpolation coefficients are determined according to the closeness of words in the semantic space. As shown by experiments, the proposed modeling and smoothing methods can significantly reduce the perplexity of language models with moderate computational cost.

Ngram language models are ubiquitous in speech applications and many other natural language systems. One issue with n-gram language models is that the language is not completely represented in the model. When words appear that are not in the model, we may need to provide a smoothing method to distribute the model probabilities over the unknown values. Many techniques exist for language model smoothing with many different performance characteristics. Often the performance of smoothing algorithms may depend on the application of the language model (so, for example, unigram models with interpolation smoothing may perform better with information retrieval applications, but trigram models with backoff smoothing might perform better for speech). This paper examines the relative performance of some selected smoothing methods with bigram language models created using chat data. The language models are used for machine translation of chat data and for creating text classification models.