A novel post-classifier for search engine using hidden Markov model

Googling for Health Information

Hidden Markov models (HMMs) are now widely used for off-line handwriting recognition in many languages and, in particular, in Arabic. As in speech recognition, they are usually built from shared, embedded HMMs at the symbol level, in which state-conditional probability density functions are modeled with Gaussian mixtures. In contrast to speech recognition, however, it is unclear which kinds of features should be used and, indeed, very different feature sets are in use today. Among them, we have recently proposed to simply use columns of raw, binary image pixels, which are directly fed into embedded Bernoulli (mixture) HMMs, that is, embedded HMMs in which the emission probabilities are modeled with Bernoulli mixtures. The idea is to bypass feature extraction and ensure that no discriminative information is filtered out during feature extraction, which in some sense is integrated into the recognition model. In this chapter, we review this idea along with some extensions that are currently providing state-of-the-art results on Arabic handwritten word recognition.

Hidden Markov Models (HMMs) is one of the sophisticated methods for ASR system. In this work, we propose an Automatic Speech-Speaker Recognition (ASSR) system based on syllable models with HMM for speech expert on a FPGA platform. HMM is used to train the acoustic model for each phoneme evaluated on two databases, i.e. THCHS-30 (Tsinghua Chinese 30 hour database) and CMU ARCTIC Databases. The binary halved clustering (BHC) method which used binary-halved splitting to generate speaker models for low complexity requirement on speaker expert. The experiments are conducted on two types of platforms, including PC and Xilinx Spartan-6 FPGA. The experimental results indicate that the proposed work can achieve 90.8% accuracy rate at Mandarin speech recognition and 86.6% accuracy rate on English speech recognition, respectively. In addition, the work also evaluates OOV (Out of Vocabulary) which reaches 88.7% of detection rate on Mandarin and 84.9% of OOV detection rate on English as well. The speaker recognition rate also reaches to 81.3% and OOS (Out of Speaker) detection rate is 80.8%, respectively.

Hidden Markov Models (HMMs) are an useful and widely utilized approach to the modeling of data sequences. One of the problems related to this technique is finding the optimal structure of the model, namely, its number of states. Although a lot of work has been carried out in the context of the model selection, few work address this specific problem, and heuristics rules are often used to define the model depending on the tackled application. In this paper, instead, we use the notion of probabilistic bisimulation to automatically and efficiently determine the minimal structure of HMM. Bisimulation allows to merge HMM states in order to obtain a minimal set that do not significantly affect model performances. The approach has been tested on DNA sequence modeling and 2D shape classification. Results are presented in function of reduction rates, classification performances, and noise sensitivity.

 Abstract—The web search engines are used to extract query specific information from this massive pool of WWW. A large number of different search engines are available to the user to satisfy their needs. Every search engine uses its own specific algorithm to rank the list of web pages returned by the search engine for the users query, so that the most relevant page appears first in the list. Users think which search engine should be selected for searching corresponding to any query topic for efficient search. For decision making on the basis of search result users want to know, whether they are significantly different or not. In this regard, a study of three popular different search engines (i.e. Google, Yahoo and Altavista) on twenty-four different query topics in terms of the quality of most relevant web page is done. The three parameters of the quality of the most relevant web page taken into consideration are: 'in depth coverage', 'ideation and clarity', and 'number of links of related articles to query topic'. For each query topic, the first web page returned by these three search engines were evaluated and graded by the experts of that specific area. Based on these grades, a table is maintained showing the relative performance of selected three search engines in terms of three parameters of quality of first returned web page and thus comparing the ranking algorithms used by them. We analyzed the result using analysis of variance and F test. By this analysis we found grade for first return page result corresponding to each and every topic through all these three search engines (Google, Yahoo and Altavista) does not vary significantly. Thus they are equivalent. So users can treat these three poular search engines as equivalent search engine for the first return page result on the basis of these three parameters.

Automatic processing of websites is of great importance for applications such as search engine that extract information from web pages. Search engines use meta tag values when classifying pages of websites. Meta tag names can change for different languages. For example, for login page, entries such as login, login page or giris, giris sayfasi may change from language to language. When the websites are examined, it can be seen that each of the pages created for the same purpose has similar designs. In this study, a deep learning based model was proposed for functional classification of web pages, regardless of language. Transfer learning was used to reduce the cost during the feature extraction process from recorded web page images. Finally, the results of two different experiments are presented for show the effectiveness of the proposed method in the classification of web pages according to their functions.

Hidden Markov models are a generalization of the finite mixture model that allow for dependence in a sequence of observations via Markovian dependence of an unobserved (hidden) sequence of states. Inference problems associated with hidden Markov models include parameter estimation and restoration of the hidden state sequence. The recursive algorithms employed to solve these problems are related to the Kalman filter and have spawned a wide variety of applications. Areas of application range from speech recognition and signal processing through analysis of molecular sequence data.

World Wide Web (WWW) is a platform that explores a wide range of information used for the development of web applications. Some examples of these applications include social network analysis, personalized item recommendations, and web page classification and ranking. Among these applications, search engines and web page ranking are particularly important as they consistently index and store billions of web pages on the internet. The main objective of this paper is to create an innovative framework for the classification and re-ranking of web pages using intelligent techniques. The framework is structured into two key phases: classification and re-ranking-based retrieval. In the initial classification phase, a series of pre-processing steps are implemented, including the elimination of HTML tags, punctuation, stop words, and the application of stemming. After pre-processing, a word-to-vector conversion is performed, followed by feature extraction utilizing Principal Component Analysis (PCA). This sequence of actions leads to optimal feature selection, which is vital for the precise classification of web pages. Given the multitude of features present in web pages that can compromise classification accuracy, this study employs a novel meta-heuristic algorithm, the Opposition Based-Tunicate Swarm Algorithm (O-TSA), to facilitate optimal feature selection. The refined features are subsequently processed through the Enhanced Convolutional-Recurrent Neural Network (E-CRNN), enhanced by O-TSA, resulting in the effective classification of diverse web page categories. In the second phase, the re-ranking process is executed using O-TSA, which establishes the objective function based on a similarity function (correlation) for URL matching, leading to optimal re-ranking of web.

Hidden Markov models are now used in many fields, for example, speech recognition, natural language processing etc. However, the mathematical foundation of analysis for the models has not yet been constructed, since the HMMs are non-identifiable. In recent years, we have developed the algebraic geometrical method that allows us to analyze the non-regular and non-identifiable models. In this paper, we apply this method to the HMM and reveal the asymptotic order of its stochastic complexity in the mathematically rigorous way.

Hidden Markov models (HMMs) have been widely used in various fields, including image categorization and retrieval. Most of the existing methods train HMMs by low-level features of image blocks; however, the blockbased features can not reflect high-level semantic concepts well. This paper proposes a new method to train HMMs by region-based features, which can be obtained after image segmentation. Our work can be characterized by two key properties: (1) Region-based HMM is adopted to achieve better categorization performance, for the region-based features accord with the human perception better. (2) Multi-layer semantic representation (MSR) is introduced to couple with region-based HMM in a long-term relevance feedback framework for image retrieval. The experimental results demonstrate the effectiveness of our proposal in both aspects of categorization and retrieval.

Hidden Markov Models (HMM) are used in handwritten strokes recognition task. The two design parameters of HMM are the number of states and number of mixtures in each state. There are two approaches for finding the number of states, namely, equal number of states and variable number of states. Since the shape of strokes will be different, variable number of states approach should be beneficial. This work proposes a curvature point detection based method to predict variable number of states for modeling a handwritten stroke. The proposed method selects appropriate points from a trace so that the portion between two consecutive points is modeled as an HMM state. Accordingly, based upon handwritten stroke shape complexity, the number of appropriate points selected will change and hence the number of states assigned to the corresponding stroke. In the proposed method, the number of states is proportional to the shape complexity of the given stroke as opposed to fixed in case of brute-force. The HMM based stroke recognizer consisting of 181 distinct strokes, was trained on a set of 52,977 examples collected from approximately 100 native Assamese writers. The evaluation was done on 43,828 examples collected from same users in different sessions. The experimental results demonstrate the benefits of the proposed technique over the brute-force method, especially in case of complex shape strokes.

Hidden Markov models (HMMs) with explicit state duration density (HMM/SD) can represent the time-varying characteristics of speech signals more accurately. However, such an advantage is reduced in relatively smooth state duration densities or long bounded duration. To solve this problem, the authors propose HMMs with global path constraint (HMM/GPC) where the transition between states occur only within prescribed time slots. HMM/GPC explicitly limits state durations and accurately describes the temporal structure of speech simply and efficiently. HMMs formed by combining HMM/GPC with HMM/SD are also presented (HMM/SD+GPC) and performances are compared. HMM/GPC can be implemented with slight modifications to the conventional Viterbi algorithm. HMM/GPC and HMM/SD+GPC not only show superior performance than the conventional HMM and HMM/SD but also require much less computation. >

Hidden Markov models (HMM) have largely demonstrated their usefulness in the fields of statistics and pattern recognition, particularly for speech recognition and hand writing recognition. In the field of genetics, the same principles of statistics and probability can be applied. DNA primarily has four bases: adenine, guanine, thymine, and cytosine, which when paired together can form nucleotides. However, the length of a nucleotide chain can be uncertain. The DNA sequence constitutes the heritable genetic information in nuclei that forms the basis for the developmental programs of all living organisms. Determining the DNA sequence is therefore useful in studying fundamental biological processes, as well as in diagnostic or forensic research. In this study, we will utilize hidden Markov models (HMM) to determine DNA sequence likelihoods. A training sequence of nucleotide bases of the first 1000 bases of rice chromosomes will be used, and the transition and emission probabilities would determine a probable DNA sequence of the next 2000 bases. This sequence should be comparable to the actual sequence. However, experimentation did not show this to be the case, despite previous experiments showing otherwise. Only a fourth of a nucleotide sequence was ever classified correctly.

Hidden Markov models (HMMs) are usually learned using the expectation maximisation algorithm which is, unfortunately, subject to local optima. Spectral learning for HMMs provides a unique, optimal solution subject to availability of a sufficient amount of data. However, with access to limited data, there is no means of estimating the accuracy of the solution of a given model. In this paper, a new spectral evaluation method has been proposed which can be used to assess whether the algorithm is converging to a stable solution on a given dataset. The proposed method is designed for real-life datasets where the true model is not available. A number of empirical experiments on synthetic as well as real datasets indicate that our criterion is an accurate proxy to measure quality of models learned using spectral learning.

Hidden Markov models (HMMs) have become within a few years the main technology for online handwritten word recognition (HWR). We consider segment models which generalize HMMs, these models aim at modeling the signal at a global level rather than at the frame level and have been shown to overcome standard HMMs in their modeling ability. We propose a segment model which allows us to automatically handle different writing styles. We compare our system on the isolated character set of the UNIPEN database with a reference system and a baseline segment model.