نشریه محاسبات نرم
سال پنجم شماره 1 (پیاپی 9، بهار و تابستان 1395)

Clustering is one of the main techniques in data mining. Clustering is a process that classifies data set into groups. In clustering, the data in a cluster are the closest to each other and the data in two different clusters have the most difference. Clustering algorithms are divided into two categories according to the type of data: Clustering algorithms for numerical data and clustering algorithms for categorical data. The clustering algorithms for categorical data are more important than clustering algorithms for numerical data because of the nature and application of these data. Each of these algorithms uses different similarity measures according to the type of data (numeric or categorical). In this paper, a new clustering method is proposed for clustering by combining Overlay and Jaccard similarity measures on a hierarchical algorithm for categorical data. Overlay measure represents similarities between the data as one and zero which caused the loss of some information. Jaccard measure If used alone to measure the similarity between data set, most clusters are selected in the particular area of data collection. So in this paper a combination of the two measures are used. Experimental results show that the proposed method improves the results of clustering. Resulted improvemen is 10% on any evaluation factor in average.

Image zooming is one of the existing issues discussed in image processing in which the person is interested in enlarging a certain picture and fitting it into the visible area of the display devices. In zooming activities it is necessary that extra pixels be inserted in the information image. The addition of the image data should be suitable with the texture of the image without producing artificial blocks. In this research, by using the approximation of the best page required Pixels are estimated. To evaluate the ability of the method, the results are evaluated on several images and compared with other methods. Average of execution time for the image 512 × 512 is 13.95 seconds. The average PSNR of the original image and image zooming is 31.29 which it shows that image zooming is very similar to the original image and the proposed method has an efficient performance.

Software project management software is the most important activity in software development, because it contains the whole software development process, from beginning to end. Software cost estimation is a challenge task in the software project management. It is an old activity in computer industry from 1940s and has been developed many times. Effort, only covers part of the cost of a software project. However, it is an essential factor for determining the cost. Therefore, in researches on software cost estimates, effort estimation and cost estimation are equivalent. A Software cost estimation model is appropriate if provides the accuracy and confidence simultaneously in cost prediction before software project contract. Due to the uncertain nature of cost estimates and in order to increase the accuracy, researchers recently have focused on machine learning techniques. In this paper, we investigated the software cost estimation by machine learning approaches, also, we introduced estimation methods, criterias to assess the accuracy of proposed methods, used datasets for evaluation, and future works in this research area.

One of the stages of quality control in porcelain producing factories is sorting that do with human eyes. Machine vision , including new methods for defect detection and sorting of different products. In this study, with defects diagnosis and as a result sorting porcelain, use from linear structured light pattern, triangulation techniques and rules governing mirrors. Also, among the defects occurred on porcelain, some defects that changes geometry, perimeter and texture of dishes surface, such as distortion, drops, depression, pin-hole, Concavity of floor have examined. With radiation linear structured light pattern on porcelains surface, shoot a picture from profiles reflected. Then, for processing and features extraction, transferred pictures to computers and after diagnosis kind of defect, get a criterion to determine the degree of dishes. Then, get the degree of the quality of dishes, by considering the defined numerically table ranking. Finally, by using proposed algorithms in this study, 2250 dishes with a predetermined degree have examined. The accuracy of the Concavity floor is 97.66% and the accuracy of the pin-hole is 98.5% appointed.

An n-dimensional hypercube, Q_n, is a graph in which vertices are binary strings of length n where two vertices are adjacent if they differ in exactly one coordinate. Hypercubes and their subgraphs have a lot of applications in different fields of science, specially in computer science. This is the reason why they have been investigated by many authors during the years. Some of their subgraphs named Fibonacci cubes and Lucas cubes are very important and are useful in interconnection networks. In this paper, after introducing these cubes, we report their metric and combinatorial properties done by different authors. Then, we present some open problems that we have been encountered during our research regarding these cubes. Finally, we briefly introduce the software named Sage which is very applicable in the calculations for theses cubes in high dimensions.