In recent years, visual programming languages (VPLs), such as Scratch, have been widely used by programming beginners. Subsequently, learners often transition to text-based programming languages (TPLs), such as Java. However, a seamless transition between these two types of programming languages has not yet been effectively established. In this study, we experimentally demonstrated that students who utilized our proposed intermediate music content between learning VPL and TPL exhibited enhanced comprehension of the TPL. Our results indicate that students who engaged with intermediate content while transitioning from VPL to TPL during their high school years performed better in their first-year college programming classes compared to those who did not use such content.

In machine learning, one of the methods to solve multiclass classification problems is a framework called Error-Correcting Output Codes (ECOC), which constructs a multiclass classifier by combining a lot of binary classifiers. ECOC assigns binary codewords to each category, and the multiclass classification performance varies depending on the code. In this study, we treat each element of the codeword as a random variable and evaluate the average performance of ECOC. As a result, for$M$class classification if the number of binary classifiers is$O(\log M)$, then the average error probability of various codes approaches that of MAP estimation. We show that the important points are the ratio between the number of binary classifiers and$\log M$and the difference between the maximum posterior probability and the second highest posterior probability for the categories.

Error-correcting output coding (ECOC) is a method for constructing a multi-valued classifier using a combination of given binary classifiers. ECOC can estimate the correct category by other binary classifiers even if the output of some binary classifiers is incorrect based on the framework of the coding theory. The code word table representing the combination of these binary classifiers is important in ECOC. ECOC is known to perform well experimentally on real data. However, the complexity of the classification problem makes it difficult to analyze the classification performance in detail. For this reason, theoretical analysis of ECOC has not been conducted. In this study, if a binary classifier outputs the estimated posterior probability with errors, then this binary classifier is said to be noisy. In contrast, if a binary classifier outputs the true posterior probability, then this binary classifier is said to be noiseless. For a theoretical analysis of ECOC, we discuss the optimality for the code word table with noiseless binary classifiers and the error rate for one with noisy binary classifiers. This evaluation result shows that the Hamming distance of the code word table is an important indicator.

Consider M-valued (M$\geq$3) classification systems realized by combination of N(N$\geq\lceil\log_{2}$M$\rceil$) binary classifiers. Such a construction method is called an Error Correcting Output Code (ECOC). First, focusing on a Reed-Muller (RM) code, we derive a modified RM (mRM) code to make it suitable for the ECOC. Using the mRM code and the Hadamard matrix, we introduce a simplex code which is one of the powerful equidistant codes. Next, from the viewpoint of system evaluation model, we evaluate the ECOC by using constructive coding described above. We show that they have desirable properties such as Flexible, Elastic, and Effective Elastic as M becomes large, by employing analytical formulas and experiments.

Error-Correcting Output Coding (ECOC) is a method for constructing a multi-valued classifier using a combination of binary classifiers. The effectiveness of ECOC for multivalued classification problems has been demonstrated by many experimental evaluations. Therefore, classification performance have strongly depended on the data under consideration, and it is not clear what kind of combinations of binary classifiers have good performance. Motivated by this fact, the authors have clarified the best combination of binary classifiers that makes ECOC, assuming a situation in which each binary classifier can estimate the true posterior probability. They also have proposed a total framework for analytical evaluation when a binary classifier outputs an estimated posterior probability that approximates the true posterior probability. These studies established a framework for evaluating the theoretical performance of ECOC.Based on these findings, this study discusses the theoretical performance of ECOC from the upper bound perspective. The results showed that increasing the Hamming distance between code words can blackuce the error rate. We then evaluate various combinations of binary classifiers based on analytical evaluation.

With the development of information technology, a wide variety of data have been accumulated, and there are many methods for analyzing such data. In this study, we model the input data and the metrics between the data based on the assumption that each metric is generated from a continuous latent variable. Specifically, we assume that the input data are generated using low-dimensional latent variables and their projection matrices. We describe a method for estimating the latent variables. Because the generative model defined in this study cannot obtain the Q function analytically, we use the Monte Carlo EM algorithm to approximate the Q function and investigate an efficient parameter estimation method. Experiments using artificial data and the 20 newsgroups dataset demonstrate the effectiveness of the proposed method.

Recently, there have been several on-demand learning systems that are not restricted by learning time or place. However, in these systems, learning content is prepared in advance for each learning course, or learning content of different difficulty levels is prepared, and learners select their learning contents. In contrast to these conventional systems, many studies have been conducted on learning systems that can grasp the learning state of individual learners and provide them with most suitable learning content. We experimentally verified a method for estimating the difficulty level of a task by focusing on alpha and beta waves. However, in practice, it is not feasible to have learners wear an electroencephalograph (EEG). This study aims to discover biometric information that can be measured by a nonwearable device as an alternative to EEG for estimating the learning state.

The biometric identification system, introduced by Willems et al., is a mathematical model to identify users based on their physical features. Although the maximum rate of the number of users which are reliably dealt with in the system (identification capacity) and the exponential behavior of the average error probability (error exponents) of the legitimate users have been revealed via information theoretic approaches, optimum error exponents has not been shown when there exists a nonlegitimate user in the system. In this paper, we formally define the reliability function as the optimum error exponent for legitimate users for a given rate of the number of legitimate users and a given error exponent for the nonlegitimate users. It is shown that the reliability function can be completely characterized by the well-known random coding exponent and the hypothesis testing error exponent.

To solve multiple classification problems with $M (\geq$ 3) categories, many studies have been devoted using $N (\geq\ \lceil\log_{2}M\rceil)$ binary $(\{0,1\})$ classifiers, where these systems are known as binary Error-Correcting Output Codes (binary ECOC). As an extended version of the binary ECOC, the ternary $(\{0,\ *,\ 1\})$ ECOC have also been discussed, where ternary classifiers classify data into positive examples when the element is 1, into negative examples when the element is 0, and no classification when the element is $*$. In this paper, we discuss the ternary ECOC system from the view point of the system evaluation model based on rate-distortion function. First, we discuss a table of M code words with length N which is given by a ternary matrix W of M rows and N columns. Next, by leveraging the benchmark data for multiclass document classification which is widely used in Japan, the relationships between the probability of classification error Peand the number of the ternary classifiers N for a given M are experimentally investigated. In addition, by assuming the M-dimensional Normal distribution for a classification data model, the relationship between Peand N for a given M is also examined. Finally, we show by the system evaluation model that the ternary ECOC systems have desirable properties such as “Flexible”, “Elastic”, and “Effective Elastic”, when M becomes large.

In recent years, the paradigm of social coding in software development has attracted attention to developers all over the world, and GitHub which is a social coding tool has spread to the area like education. There are many cases using it as a platform of PBL (Project Based Learning). However, since GitHub is not a tool for education, it is difficult to evaluate learners. This research focuses on the contribution of learners and proposes a system that teachers can grasp the contribution of learners.

A decision criterion called LR+Th for decision feedback scheme was proposed by Hashimoto. Though Forney’s decision criterion (FR) is optimal and meets the Neyman-Pearson’s lemma, LR+Th is suboptimal. However, the error exponent of LR+Th is shown to be asymptotically equivalent to that of FR by random coding arguments for block codes. In this paper, applying the technique of DS2 bound, we derive an upper bound for the error probability of LR+Th with the ensemble of linear block codes. It elucidates the relation between the random coding exponents of block codes and those of linear block codes.

In coding theory, it is important to find upper bounds for the code size given a code length and minimum distance. The Hamming bounds and Linear Programming (LP) bounds were proposed in previous works. On the other hand, Masnick et al. proposed Unequal Error Protection (UEP) codes and modified Hamming bounds as upper bounds for the code size of UEP codes. In our previous work, we defined 2-level UEP codes as a subclass of UEP codes, and derived LP bounds for 2-level UEP codes. In this paper, we define multi-level UEP codes by extending 2-level UEP codes, and derive LP bounds for multi-level UEP codes. Moreover, we show that LP bounds for UEP codes are tighter upper bound than modified Hamming bounds.

Various factors affect learning, such as the quality and difficulty of the learning contents and the learner's proficiency, and it may be possible to detect their effects by using the learner's browsing and edit history as well as biological information such as brain waves and eye movement. We investigated the use of brain waves to estimate the degree of difficulty of a learning task. The first experiment using a simple typing test confirmed a previous finding that the β-wave to α-wave ratio increases with task difficulty. Furthermore, the low-β-wave to low-α-wave ratio, where "low" means low frequency, observed under various learning conditions was found to depend on the frequencies of the waves, and the value of the ratio was shown to represent task difficulty. The second experiment in which the change in brain waves was measured using a simple electroencephalograph (EEG) as the examinees became used to the task showed that the values for the examinees who reported that the task was easy fell gradually, supporting our finding that the value of the ratio represents task difficulty. This information could be used to dynamically adjust task difficulty and thereby optimize the learning effect.

In this manuscript, we investigate a collaborative filtering method to characterize consumption behavior of customers and services with various attributes for marketing. We assume that each customer and service have the invisible attribute which is called latent class. Assuming a combination of attribute values of a customer and service is classified to a latent class, furthermore, we propose a new Bayesian statistical model that consumption behavior is probabilistically arise based on a latent class combination of a customer, service and attribute values. Then, we show the method to estimate parameters of a statistical model based on the variational Bayes method and the mean field approximation. Consequently, we show the effectiveness of the proposed model and the estimation method by simulation.

In this manuscript, we investigate a collaborative filtering method to characterize consumption behavior (or evaluation) of customers (or users) and services (or items) for marketing. Assuming that each customer and service have the invisible attribute, which is called latent class, we propose a new Bayesian statistical model that consumption behavior is probabilistically arise based on a latent class combination of a customer and service. Then, we show the method to estimate parameters of a statistical model based on the variational Bayes method and the mean field approximation. Consequently, we show the effectiveness of the proposed model and the estimation method by simulation and analyzing actual data.

In this paper, we develop a new decoding algorithm of binary linear codes for symbol-pair read channel. The Symbol-pair read channel has recently been introduced by Cassuto and Blaum to model channel whose write resolution is higher than read resolution. The proposed decoding algorithm is based on the linear programming (LP). It is proved that the proposed LP decoder has the maximum-likelihood (ML) certificate property, i.e., the output of the decoder is guaranteed to be the ML codeword when it is integral. We also introduce the fractional pair distance dfpof the code which is a lower bound on the minimum pair distance. It is proved that the proposed LP decoder corrects up to ⌈dfp/2⌉ - 1 errors.

In this paper, we focus on pattern recognition based on the vector space model. As one of the methods, distance metric learning is known for the learning metric matrix under the arbitrary constraint. Generally, it uses iterative optimization procedure in order to gain suitable distance structure by considering the statistical characteristics of training data. Most of the distance metric learning methods estimate suitable metric matrix from all pairs of training data. However, the computational cost is considerable if the number of training data increases in this setting. To avoid this problem, we propose the way of learning distance metric by using the each category centroid. To verify the effectiveness of proposed method, we conduct the simulation experiment by using benchmark data.

This paper proposes a new latent class model to represent user segments in a marketing model of electric commerce sites. The aspect model proposed by T. Hofmann is well known and is also called the probabilistic latent semantic indexing (PLSI) model. Although the aspect model is one of effective models for information retrieval, it is difficult to interpret the meaning of the probability of latent class in terms of marketing models. It is desirable that the probability of latent class means the size of customer segment for the purpose of marketing research. Through this formulation, the simulation analysis to dissect the several situations become possible by using the estimated model. The impact of the strategy that we contact to the specific customer segment and make effort to increase the number of customers belonging to this segment can be predicted by using the model demonstrating the size of customer segment. This paper proposes a new model whose probability parameter of latent variable means the rate of users with the same preference in market. By applying the proposed model to the data of an internet portal site for job hunting, the effectiveness of our proposal is verified.

By introducing the technology of the virtual desktop infrastructure (VDI) to e-learning systems, the identical learning environment can be??provided to the learner at any time, and at any place, and also even if the learner stops learning or working, he can restart it by different terminals including different operating systems. On the other hand, the quality of screen images on the desktop would be affected by the quality of service (QoS) of the network, since the screen images are transferred to the client from the server by their transfer protocol. In this paper, we discuss the influence of the QoS upon the learning styles caused by the usability over the VDI. By constructing the experimental network using network emulators, we evaluate and discuss the influence of the QoS such as a round trip delay time (DT) and a bandwidth (BW) upon learning style measured by the quality of experience (QoE). As a result, we clarify the relationships between the QoS of the network and the QoE of the e-learning styles.

In this paper, we consider a completion problem of multiple related matrices. Matrix completion problem is the problem to estimate unobserved elements of the matrix from observed elements. It has many applications such as collaborative filtering, computer vision, biology, and so on. In cases where we can obtain some related matrices, we can expect that their simultaneous completion has better performance than completing each matrix independently. Collective matrix factorization is a powerful approach to jointly factorize multiple matrices. However, existing completion algorithms for the collective matrix factorization have some drawbacks. One is that most existing algorithms are based on non-convex formulations of the problem. Another is that only a few existing algorithms consider the strength of the relation among matrices and it results in worse performance when some matrices are actually not related. In this paper, we formulate the multiple matrix completion problem as the convex optimization problem. Moreover, it considers the strength of the relation among matrices. We also develop an optimization algorithm which solves the proposed problem efficiently based on the alternating direction method of multipliers (ADMM). We verify the effectiveness of our approach through numerical experiments on both synthetic data and real data set: MovieLens.

F. P. Preparata et al. proposed a fault diagnosis model (PMC model) to find all fault units in the multicomputer system by using outcomes that each unit tests some other units. T. Kohda proposed a highly structured(HS) system and the syndrome analysis method(SAM) to diagnose from local testing results. In this paper, we introduce the maximum a posteriori probability algorithm(MAPDA) for the HS system in the probabilistic fault model. Analyzing the MAPDA, we show that the SAM is closer to the MAPDA as the fault probability becomes smaller. Finally, we show the robustness of the SAM in the HS system.

In this paper, we focus on pattern recognition based on the vector space model with the high dimensional and sparse data. One of the pattern recognition methods is metric learning which learns a metric matrix by using the iterative optimization procedure. However most of the metric learning methods tend to cause overfitting and increasing computational time for high dimensional and sparse settings. To avoid these problems, we propose the method of l1regularized metric learning by using the algorithm of alternating direction method of multiplier (ADMM) in the supervised setting. The effectiveness of our proposed method is clarified by classification experiments by using the Japanese newspaper article and UCI machine learning repository. And we show proposed method is the special case of the statistical sparse covariance selection.

Experiments usually aim to study how changes in various factors affect the response variable of interest. Since the model used most often at present in experimental design is expressed through the effect of each factor, it is easy to understand how each factor affects the response variable. However, since the model contains redundant parameters, a considerable amount of time is often necessary to implement the procedure for estimating the effects. On the other hand, it has recently been shown that the model in experimental design can also be expressed in terms of an orthonormal system. In this case, the model contains no redundant parameters. Moreover, the theorem with respect to the sum of squares for the 2-factor interaction, needed in the analysis of variance (ANOVA) has been obtained. However, 3-factor interaction is often to be considered in real cases, but the theorem with respect to the sum of squares for the 3-factor interaction has not been obtained up to now. In this paper, we present the theorem with respect to the sum of squares for the 3-factor interaction in a model based on an orthonormal system. Furthermore, we can also obtain the theorem for interactions with 4 or more factors by the similar proof. Hence, in any real case, we can execute ANOVA in the model based on an orthonormal system.

In this paper, a new attack model in which the number of colluders are distributed according to a certain probability distribution is introduced. Two classes of collusion attacks which include well-known collusion attacks in the context of multimedia fingerprinting are provided. For these two attack classes, achievable rates for the unknown size of the actual colluders are derived. Based on the derived achievable rates, achieve rates for some particular attacks are investigated. For the AND attack, the bound derived in this paper coincides with the previous known bound, although the attack model in this paper does not assume that the decoder knows the actual number of colluders. Moreover, for the averaging attack, it is clarified that derived achievable rate is larger than previously known bound with random linear codes.

The SQL injection attack causes very serious problem to web applications which have database including personal data. To detect the SQL injection attack, the parsing and the black list based on the existed attack have been widely used. However, a new attack method that is not included by the black list has still been developed. In this paper, we proposed the detecting method based on single character, and show the effectiveness of the proposed method experimentally using both attack and normal samples.

We discuss disk allocation methods for Cartesian product files by introducing error correcting codes, and have clarified the performance of the methods by system evaluation models developed by using rate distortion theory. Let us assume qnCartesian product files with n attributes and q actual values in each attribute, and store qnfiles into G(≤ qn) disks. For a partial match access request, we represent new disk allocation methods which able to access the disks in parallel as much as possible, where the partial match access request includes an indefinite case (don't care: “*”) in some attributes and the * requires to access the files with corresponding to the attribute for the all actual attribute values. In this paper, we propose to apply unequal error protection codes to the case where the probabilities of occurrence of the * in the attributes for a partial match access request are not the same. We show the disk allocation methods have desirable properties as n becomes large.

F.P.Preparata et al. have proposed a fault diagnosis model to find all faulty units in the multicomputer system by using outcomes which each unit tests some other units. In this paper, for probabilistic diagnosis models, we show an efficient diagnosis algorithm to obtain a posteriori probability that each of units is faulty given the test outcomes. Furthermore, we propose a method to analyze the diagnostic error probability of this algorithm.

The sigmoid function has been widely used in the problem of the two value distinction. In this paper, we prepare a function which is similar to a sigmoid function, and show that our proposed model is valid to the SQL Injection attack detection than the method using a sigmoid function.

Allocation methods for Cartesian product files on multiple disks using linear error-correcting codes are discussed. In this paper, we propose an allocation method using unequal error protection (UEP) codes. Codewords of an UEP code have some special bits which are protected against a greater number of errors than the other bits. We firstly assume a model that “*”, which means “don't care”, appears with different probability in each attribute of queries. In this case, the average access time can be calculated using the split distance distribution. Finally, we illustrate the average access time of the methods using UEP codes.

Experiments usually aim to study how changes in various factors affect the response variable of interest. Since the response model used most often at present in experimental design is expressed through the effect of each factor, it is straightforward to ascertain how each factor affects the response variable. However, since the response model contains redundant parameters, we must calculate the degrees of freedom defined by the number of independent parameters in the analysis of variance. In this paper, we show that through a description of experimental design based on an orthonormal system, the response model can be expressed using only independent parameters. Hence, we do not have to calculate the degrees of freedom defined by the number of independent parameters.

The smart phone penetration rate has increased recently. There are also many commercial terminals for receiving network services. When a service is received by a smart phone or PC, user authentication is very important for ensuring secure and safe use. Currently, authentication is required each time a user changes the terminal on which a service is received. We propose a system that uses a smart phone as a storage device for authentication information, such as ID, password, and cookie information. With this system, the smart phone and various terminals cooperate through short distance wireless telecommunications technologies such as Bluetooth. We evaluate performance of our proposed system. As a result, the user can input authentication information to and receive a service on a terminal by simply swiping the smart phone over the terminal.

In our previous paper, we proposed a new classification technique called the Frequency Ratio Accumulation Method (FRAM). This is a simple technique that adds up the ratios of term frequencies among categories, and it is able to use index terms without limit. Then, we adopted the Character N-gram to form index terms, thereby improving FRAM. However, FRAM did not have a satisfactory mathematical basis. Therefore, we present here a new mathematical model based on a “Vector Space Model” and consider its implications. The proposed method is evaluated by performing several experiments. In these experiments, we classify newspaper articles from the English Reuters-21578 data set, a Japanese CD-Mainichi 2002 data set using the proposed method. The Reuters-21578 data set is a benchmark data set for automatic text categorization. It is shown that FRAM has good classification accuracy. Specifically, the micro-averaged F-measure of the proposed method is 92.2% for English. The proposed method can perform classification utilizing a single program and it is language-independent.

The context tree models are widely used in a lot of research fields. Patricia like trees are applied to the context trees that are expanded according to the increase of the length of a source sequence in the previous researches of non-predictive source coding and model selection. The space complexity of the Patricia like context trees are O(t) where t is the length of a source sequence. On the other hand, the predictive Bayes source coding algorithm cannot use a Patricia like context tree, because it is difficult to hold and update the posterior probability parameters on a Patricia like tree. So the space complexity of the expanded trees in the predictive Bayes coding algorithm is O(t2). In this paper, we propose an efficient predictive Bayes coding algorithm using a new representation of the posterior probability parameters and the compact context tree holding the parameters whose space complexity is O(t).

In this paper, we generalize the sampling theorem for bandlimited functions over the Boolean domain to a sampling theorem for bandlimited functions over a GF(q)ndomain. We also present a theorem for the relation between the parity check matrix of a linear code and any distinct error vectors. Lastly, we clarify the relation between the sampling theorem for functions over a GF (q)ndomain and linear codes.

This paper introduces two channel models of correlated parallel channels. Then we analyze structure of error correcting codes over these correlated parallel channels. We derive necessary and sufficient conditions for these codes and some code construction is presented. We also show some upper and lower bounds on the coding rate of the error correcting codes for correlated parallel channels. The introduced channel models are related to burst error channels and the codes analyzed in this paper can be used as asymmetric interleaving codes for burst error channels.

In this paper, transmission schemes with noiseless and high capacity feedback channel is considered. We propose two types of transmission schemes using LDPC codes and clarify the density evolution analysis method for these proposed schemes. We investigate the performance of the proposed schemes by the density evolution analysis and computer simulations. The result shows some interesting characteristics for schemes with high capacity feedback channel.

In this paper, we generalize the achievability of variable-length coding from two viewpoints. One is the definition of an overflow probability, and the other is the definition of an achievability. We define the overflow probability as the probability of codeword length, not per symbol, is larger thanetanand we introduce theisin-achievability of variable-length codes that implies an existence of a code for the source under the condition that the overflow probability is smaller than or equal toisin. Then we show that theisin-achievability of variable-length codes is essentially equivalent to theisin-achievability of fixed-length codes for general sources. Moreover we show the condition ofisin-achievability for some restricted sources givenisin.

By combining statistical analyses and information retrieval techniques, an efficient way for knowledge discovery from questionnaires is discussed. Since usual questionnaires include questions answered by a fixed format and those by a free format, it is important to introduce the methods by both data mining and text mining. The answers by the fixed format are called "items", and those by the free format, simply "texts". In this paper, using an algorithm for processing answers with both the items and the texts and that for extracting important sentences from texts combined with statistical techniques, a method for analyzing the questionnaires is established. The method is applied to a case of improvements for the quality of education by which the student questionnaire is executed to a class, and we obtain useful knowledge which leads to faculty development.

In the present paper, we consider the automatic text categorization as a series of information processing and propose a new classification technique called the Frequency Ratio Accumulation Method (FRAM). This is a simple technique that calculates the sum of ratios of word frequency in each category. However, in FRAM, feature terms can be used without limit. Therefore, we propose the use of the character N-gram and the word N-gram as feature terms using the above-described property of FRAM. Next, we evaluate the proposed technique through a number of experiments. In these experiments, we classify newspaper articles from Japanese CD-Mainichi 2002 and English Reuters-21578 using the Naive Bayes method (baseline method) and the proposed method. As a result, we show that the classification accuracy of the proposed method is far better than that of the baseline method. Specifically, the classification accuracy of the proposed method is 87.3% for Japanese CD-Mainichi 2002 and 86.1% for English Reuters-21578. Thus, the proposed method has very high performance. Although the proposed method is a simple technique, it provides a new perspective and has a high potential and is language-independent. Thus, the proposed method can be expected to be developed further in the future.

Digital fingerprinting, a copyright protection technique for digital contents, is considered. Digital fingerprinting should deter collusion attacks, where several fingerprinted copies of the same content are mixed to disturb their fingerprints. In this paper, we consider the averaging attack, which has effect for multimedia fingerprinting. We propose new collusion-secure fingerprinting codes based on finite geometries (FGs) which increase the rate of conventional collusion-secure codes, while they guarantee to identify the same number of colluders. Due to the new FG-based fingerprinting codes, the system can deal with a larger number of users to distribute a digital content.

The problem of universal codes with side information is investigated from Bayes criterion. We propose side information context tree models which are an extension of context tree models to sources with side information. Assuming a special class of the prior distributions for side information context tree models, we propose an efficient algorithm of Bayes code for the models. The asymptotic code length of the Bayes codes with side information is also investigated

The best-known algorithm for the decoding of low-density parity-check (LDPC) codes is the sum-product algorithm (SPA). The SPA is a message-passing algorithm on a graphical model called a factor graph (FG). The performance of the SPA depends on a structure of loops in a FG. Pearl showed that loops in a graphical model could be erased by the clustering method. This method clusters plural nodes into a single node. In this paper, we show several examples about a decoding on a FG to which the clustering method is applied. And we propose an efficient decoding algorithm for it. For a binary erasure channel (BEC), the performance with this method goes up clearly.

In this paper, we consider the Bayesian approach for representation of a set of documents. In the field of representation of a set of documents, many previous models, such as the latent semantic analysis (LSA), the probabilistic latent semantic analysis (PLSA), the semantic aggregate model (SAM), the Bayesian latent semantic analysis (BLSA), and so on, were proposed. In this paper, we formulate the Bayes optimal solutions for estimation of parameters and selection of the dimension of the hidden latent class in these models and analyze it's asymptotic properties.

Based on techniques in information retrieval, we discuss the methods for knowledge acquisition from the documents composed of both fixed and free formats. The documents with the fixed format imply items with those selected from the sentences, words, symbols, or numbers, while the documents with free format are with the usual text. In this paper, starting with the item-document matrix and term-document matrix used for the representation of a document set, we propose a new method for knowledge acquisition taking simultaneously into account of both fixed and free formats. A method based on the probabilistic latent semantic indexing (PLSI) model is used for clustering a set of documents. The proposed method is applied to a document set given by the questionnaires of students taken for the purpose of faculty development. We show the effectiveness of the proposed method compared to the conventional method.

In this paper, we discuss the difference in code lengths between the code based on the minimum description length (MDL) principle (the MDL code) and the Bayes code under the condition that the same prior distribution is assumed for both codes. It is proved that the code length of the Bayes code is smaller than that of the MDL code by o(1) or O(1) for the discrete model class and by O(1) for the parametric model class. Because we can assume the same prior for the Bayes code as for the code based on the MDL principle, it is possible to construct the Bayes code with equal or smaller code length than the code based on the MDL principle. From the viewpoint of mean code length per symbol unit (compression rate), the Bayes code is asymptotically indistinguishable from the MDL two-stage codes.

Discusses a problem of deduction with uncertainty that has been dealt with by various diagnostic expert systems. First, we propose a mathematical framework of deductive reasoning with uncertainty. The subject of the reasoning is the calculation of conditional probabilities. Second, we establish a new reasoning method. Our deduction algorithm can compute the conditional probabilities precisely. To put it another way around, the result minimizes the divergence.

A new soft-decision maximum-likelihood decoding algorithm is proposed, which generates a set of candidate codewords using hard-decision bounded-distance decoding. By improving the generating method of input vectors for the bounded-distance decoding due to Kaneko et al. (see ibid., vol.40, no.3, p.320-27, 1994), the decoding time complexity is reduced without degradation of the performance. The space complexity is dependent on the bounded-distance decoding.

A new soft decoding algorithm for linear block codes is proposed. The decoding algorithm works with any algebraic decoder and its performance is strictly the same as that of maximum-likelihood-decoding (MLD). Since our decoding algorithm generates sets of different candidate codewords corresponding to the received sequence, its decoding complexity depends on the received sequence. We compare our decoding algorithm with Chase (1972) algorithm 2 and the Tanaka-Kakigahara (1983) algorithm in which a similar method for generating candidate codewords is used. Computer simulation results indicate, for some signal-to-noise ratios (SNR), that our decoding algorithm requires less average complexity than those of the other two algorithms, but the performance of ours is always superior to those of the other two.>

Considering inductive inference and deductive inference as not individual processes but a serial process of information processing, the serial inference procedure is studied for two purposes: the compression of observed facts and the prediction of new well-formed formulas. A serial inference process scheme that uses the correspondence to source coding and prediction problems is proposed. The optimal inference procedures for the two purposes are shown in the proposed scheme.>

The problem of distortionless encoding when the parameters of the probabilistic model of a source are unknown is considered from a statistical decision theory point of view. A class of predictive and nonpredictive codes is proposed that are optimal within this framework. Specifically, it is shown that the codeword length of the proposed predictive code coincides with that of the proposed nonpredictive code for any source sequence. A bound for the redundancy for universal coding is given in terms of the supremum of the Bayes risk. If this supremum exists, then there exists a minimax code whose mean code length approaches it in the proposed class of codes, and the minimax code is given by the Bayes solution relative to the prior distribution of the source parameters that maximizes the Bayes risk.>

The operating cost of distributed database systems (DDBS) usually depends on the storage costs of file copies, i.e file redundancy, and the communication cost of queries and for updates. A basic model is proposed to clarify the tradeoffs between the file redundancies and the communication cost of queries from the viewpoint of rate-distortion theory and its applications, for the case in which almost all requests are assumed to be queries. The result suggests that the addition of a few file copies is effective for the bus or ring network subsystems and the existence of only one file is essential to the star network subsystem, as the system size becomes large. The model leads to a useful measure for selecting the file-allocation design of a large scale DDBS.>

In information retrieval systems, the consecutive retrieval (CR) property enables the design of the system with a minimal search time and no redundant storage. However, the CR property cannot exist between every arbitrary query set and every record set unless we relax these restrictions. We use the variable-buffer quasi-CR (V-QCR) property scheme which tolerates variable buffer size in the main storage and several secondary storages. In secondary file organization, we propose two subschemes: 1) the relaxed CR (RCR) property subscheme which tolerates the duplication of records and multiple access, and 2) the CR property with dummy records (CR-D property) subscheme which tolerates the dummy records and the variable buffer size. For each proposed subscheme, we evaluate the trade-offs between the system redundancy and the loss of search time by rate-distortion theoretic approach. The result indicates that it is worthwhile to search for algorithms that will generate the secondary storage locations satisfying RCR and CR-D property subschemes, when the system size becomes sufficiently large.

By modifying product codes, a new coding scheme and its decoding method are proposed. Compared to a product codeA, the first stage codeA_{1}of the new codeA_{M}is constructed in the same way as that of the codeAexcept that it has at least one subcode, while the second stage codesA_{2}^{(j)}of the codeA_{M}are a set of codes With the same length and different rates. The new coding scheme has a smaller upper bound on the probability of decoding error than the original product coding scheme for any given nonzero rate less than the capacity of a binary symmetric channel. An example is given for which the rate is increased compared With the original product code, at a fixed probability of decoding error, for a relatively short code length.

Question-answering systems have been studied from the viewpoint of achieving drastic savings of storage with a small error tolerance. A system is called elastic if a drastic memory saving may be achieved by tolerating a low level of indefiniteness of the answer. A necessary condition for the system to be elastic is that the query sets exhibit diverging redundancy. In the binary-valued retrieval problems discussed by P. Elias and R.A. Flower (1975), a system that admits the set of all binary questions regarding an arbitrary binary data with the highest redundance has been shown to be inelastic. On the other hand, a system that admits the exact match question set regarding an arbitrary binary data with the second highest redundancy is found to be elastic.

The lower bound m the error exponent for the generalized version of concatenated codes is shown to be improved at low rates for binary-input memoryless channels.

New coding and decoding schemes based on concatenated codes are proposed. The new coding schemes are potentially superior to Forney's original concatenation scheme in the sense that for discrete memoryless channels the former has a smaller upper bound on the probability of decoding error for the same order of decoding complexity.

A family of codes of lengthn=q^{s+l}over GF(q), with2s \leq qare presented which are constructed by superimposing concatenated codes on a concatenated code. The raterand the distance ratio\deltaof the new codes satisfy the relationr=1-\delta+\delta \ln (\delta)for sufficiently large values ofnandq/s. The new codes are superior to the comparable Bose-Chaudhuri-Hocquenghem (BCH) codes, fors\geq 3, in the sense that they contain more codewords. An asymptotically good code constructed using these new codes has a distance ratio greater than those of other asymptotically good codes known to the authors for rates smaller than 0.007.

A new class of asymptotically good codes is constructed by iteratively superimposing Justesen codes on a Justesen code. The lower bound for the new codes lies above the Zyahlov bound for rater, 0.31 < r < 1.

We present new classes of binary codes that are constructed on the basis of concatenated codes and product codes. We discuss the random-error-correction capabilities of these codes. Some examples of the codes for the correction of random errors are given which have at least as many codewords as the best codes previously known (to the authors) with the same minimum distance and same number of check symbols. The burst-error-correction capabilities of the codes are also discussed. Several examples of the codes for the correction of both random errors and burst errors are given. A decoding algorithm for the codes is also described.

An erasures-and-errors decoding algorithm for Goppa codes is presented. Given the Goppa polynomial and the modified syndrome polynomial, a modified key equation is solved using Euclid's algorithm to determine the error locator polynomial and the errata evaluator polynomial.

It is shown that Goppa codes with Goppa polynomial\{g(z)\}^{q}have the parameters: lengthn \leq q^{m} - s_{o}, number of check symbolsn - k \leq m (q - 1) (\deg g), and minimum distanced \geq q (\deg g) + 1, whereqis a prime power,mis an integer,g(z )is an arbitrary polynomial overGF(q^{m}), and so is the number of roots ofg(z)which belong toGF(q^{m}). It is also shown that all binary Goppa codes of lengthn \leq 2^{m} - s_{o}satisfy the relationn - k \leq m (d - 1)/2. A new class of binary codes withn \leq 2^{ m} + ms _{0}, n - k \leq m (\deg g) + s_{0}, andd \leq 2(\deg g) + 1is constructed, as well as another class of binary codes with slightly different parameters. Some of those codes are proved superior to the best codes previously known. Finally, a decoding algorithm is given for the codes constructed which uses Euclid's algorithm.

In this correspondence, we present a new class of binary codes that are constructed on the basis of BCH codes. Some examples of these codes are given, having more codewords than the best codes previously known (to the authors) with the same minimum distance and number of check symbols. A decoding algorithm for the codes is also described.

In this correspondence, we construct a new class of binary codes by exploiting the symmetry properties of the parity check matrix of the Srivastava codes. The construction is a generalization of Goppa's construction [1]. A number of the binary codes constructed are proved equal, or superior, to the best codes previously known.