Kin-ichiroh Tokiwa

Generally, the majority of customers move along the areas located outside of the main circular path on the sales floor of a supermarket. At the same time, statistically it has been determined that a longer shopping path leads to a higher purchase volume and higher volume customers have the possibility of spending more time within the inner part of a sales floor. Thus, we expect an improvement in purchasing by opening shopping paths into the inner areas to enhance customers' chances to discover new products. Realising an effective path extension to increase customers' purchases requires us to perceive the features of higher and lower volume customers. The study aims to establish a classification rule based on customers' behavioural features within inner areas. The paper provides an explanation of our method to quantify customers' behavioural features and reports the outcomes from the exploratory analysis using real-world shopping path and purchase data.

Many university students, especially engineering students, hesitate to state their opinions or ask questions in front of other students during classroom lectures since they do not want to feel embarrassed or humiliated, or just because they need some practice speaking in front of an audience. In this study, we develop a gamified (arranged as a game), half-anonymized (incorporating only very slight clues to identify other users) education support system that can be utilized to improve classroom interactions and propose an educational methodology that encourages class members to engage simultaneously in both off-and online communication. Using the system, students take part in an online, mediated class-wide discussion, while also engaging in face-to-face verbal discussion with their neighbors. The system provides students with an interactive, cooperative learning environment free of the influence of real-world human relationships that can impede students' active engagement in class. The contribution of the paper is in its investigation of how to make students more proactive in class by harnessing the power of gamification and limited anonymity. We also report the findings from an experimental implementation of the system in a class of 17 university engineering students. © 2013 IEEE.

This paper describes a method of readability assessment for web documents. Readability is the ease in which text can be read and understood. We hypothesize that the readability is determined whether a reader can easily grasp text structures. The impression and the complexity of text are significant factors. We extract the features about impression and complexity from a plain text and additional data such as HTML tags.<br>In order to compare effect of extracting features, we are assessing readability rank by machine learning. We conduct 5-fold cross validation for each domain, and calculate the root mean squared error between the actual rank and the estimated rank. The cross validation experiments confirm that the performance of our method is high measured. It shows effectiveness of extracting features about the impression and the complexity for readability assessment.

By utilizing the method to get the upper bound on the average probability of undetected error for all binary linear systematic block codes, we newly get each upper bound on the average probability of undetected error for the ensembles of all binary product codes and all binary expansions of concatenated codes which include the constructive algebraic codes satisfying Shannon's channel coding theorem or being the asymptotically good. By comparing those upper bounds, we can clearly show that the average capability of concatenation structure is stronger than that of product codes, although the bound for concatenated codes is below that for the ensemble of all binary linear systematic block codes. © 2010 IEEE.

Recently, Schwartz and Etzion proposed several efficient code construction methods of two-dimensional linear codes capable of correcting a single arbitrary cluster of errors of size b. Some of their methods, however, impose the constraint that the degree m of GF (2(m)) must be even. In this correspondence, we investigate whether such Schwartz-Etzion's methods are applicable to the case in is odd. We provide necessary and sufficient conditions for the original Schwartz-Etzion's methods to be used for constructing two-dimensional cluster-error-correcting linear codes for odd m, and show that the methods are available extensively.

Quantum error-correcting codes have been developed as one of the promising tools for protecting quantum information against quantum errors. A great deal of effort has been made mainly to construct efficient quantum random-error-correcting codes. In this article, we investigate a class of quantum codes capable of correcting quantum burst errors, and present a list of some new good quantum burst-error-correcting codes of length less than or equal to 51.

Justesen first constructed asymptotically good concatenated codes such that the outer code is Reed-Solomon (RS) code and the inner code is Wozencraft's ensemble of randomly shifted codes. Kolev gave the weight distributions of Justesen codes in some cases. In this paper, we give the weight distributions of Justesen codes obtained by using some other RS codes as the outer code.

Recently, Vatan, Roychowdhury and Anantram have presented two types of revised versions of the Calderbank-Shor-Steane code construction, and have also provided an exhaustive procedure for determining bases of quantum error-correcting codes. In this paper, we investigate the revised versions given by Vatan ct al., and point out that there is no essential difference between them. In addition, we propose an efficient algorithm for searching for bases of quantum error-correcting codes. The proposed algorithm is based on some fundamental properties of classical linear codes, and has much lower complexity; than Vatan et at's procedure.

When unidirectional errors and symmetric errors occur in the channel, it is well known that codes capable of correcting up to t symmetric errors and detecting all t + 1 and greater unidirectional errors, i.e., t-error correcting/all unidirectional error detecting (t-EC/AUED) codes, are effective. Furthermore, constant weight codes with t-error correcting capability are known as t-EC/AUED codes with high coding rate. Recently, Bose and Rao have proposed a single error correcting constant weight code with high coding rate that is realizable with relatively simple encoding and decoding operations. In this paper, a new construction method for single error correcting constant weight codes is proposed. The encoding and decoding operations are not too complex, and it is possible to achieve coding rates even higher than those of the Bose-Rao codes. Furthermore, with the construction method proposed in this paper, not only in comparisons restricted to constant weight codes, but also in comparisons to previous codes that are known as high rate codes, it is possible to construct codes with the same error correcting capability that have even higher coding rates. (C) 1997 Scripta Technica, Inc.

The authors previously considered the problem of deriving the complete weight distribution (CWD) of Reed-Solomon (RS) codes and developed a formula which established the relationship between a specific codeword set and the CWD of an RS code. If this formula is used, then the CWD of an RS code can be determined by examining fewer codewords than would be required by an exhaustive search. However, subsequent investigation showed that this formula can be applied in its current form only to codes with prime length and not to codes with composite length. A new form of this formula given herein does not depend on whether the code length is prime or composite. The set of codewords that must be examined to determine the CWD of an RS code is respecified and a formula that gives the relationship between this newly specified codeword set and the CWD of an RS code is derived. It is also shown that, when the generator polynomial of an RS code does not have unity as a root, then the number of codewords that must be examined can be reduced even further.

Coding scheme is discussed for M-Choose-T communication in which at most T active users out of M potential users simultaneously transmit their messages over a common channel. The multiple-access channel considered in this paper is assumed to be a time-discrete noiseless adder channel without feedback with T binary inputs and one real-valued output, and is used on the assumption of perfect block and bit synchronization among users. In this paper a new class of uniquely decodable codes is proposed in order to realize error-free M-Choose-T communication over the adder channel described above. These codes are uniquely decodable in the sense that not only the set of active users can be specified but also their transmitted messages can be recovered uniquely as long as T or fewer users are active simultaneously. It is shown that these codes have a simple decoding algorithm and can achieve a very high sum rate arbitrarily close to unity if exactly T users are active.