Toward a classification of discourse patterns in asynchronous online discussions

The goal of this study was to develop a classification for a range of discourse patterns that occur in text-based asynchronous discussion forums, and that can aid in the distinction of three modes of discourse: knowledge sharing, knowledge construction, and knowledge building. The dataset was taken from Knowledge Forum® databases in the Knowledge Building Teacher Network in Hong Kong, and included three discussion views created for different classes: Grade 5 Science, Grade 10 Visual Arts, and Grade 10 Liberal Studies. We used a combination of qualitative coding and narrative analysis as well as teachers’ understanding of online discourse to analyze student discussions. Nine discourse patterns were identified. These patterns revealed a variety of ways in which students go about their collaborative interactions online and demonstrated how and why students succeed or fail in sustaining collaborative interactions. This study extended the three modes of online discourse and developed different discourse patterns, which are efforts to provide instructional guidance. The implications of supporting productive discourse and the enactment of CSCL innovations in classrooms are discussed.     

Hybrid immune algorithm based on greedy algorithm and delete-cross operator for solving TSP

This paper first introduces the fundamental principles of immune algorithm (IA), greedy algorithm (GA) and delete-cross operator (DO). Based on these basic algorithms, a hybrid immune algorithm (HIA) is constructed to solve the traveling salesman problem (TSP). HIA employs GA to initialize the routes of TSP and utilizes DO to delete routes of crossover. With dynamic mutation operator (DMO) adopted to improve searching precision, this proposed algorithm can increase the likelihood of global optimum after the hybridization. Experimental results demonstrate that the HIA algorithm is able to yield a better solution than that of other algorithms, which also takes less computation time.     

Band selection for hyperspectral images using probabilistic memetic algorithm

Band selection plays an important role in identifying the most useful and valuable information contained in the hyperspectral images for further data analysis such as classification, clustering, etc. Memetic algorithm (MA), among other metaheuristic search methods, has been shown to achieve competitive performances in solving the NP-hard band selection problem. In this paper, we propose a formal probabilistic memetic algorithm for band selection, which is able to adaptively control the degree of global exploration against local exploitation as the search progresses. To verify the effectiveness of the proposed probabilistic mechanism, empirical studies conducted on five well-known hyperspectral images against two recently proposed state-of-the-art MAs for band selection are presented.     

Image inpainting algorithm based on TV model and evolutionary algorithm

With the development of modern image processing techniques, the numbers of images increase at a high speed in network. As a new form of visual communication, image is widely used in network transmission. However, the image information would be lost after transmission. In view of this, we are motivated to restore the image to make it complete in an effective and efficient way in order to save the network bandwidth. At present, there are two main methods for digital image restoration, texture-based method and non-textured-based method. In the texture-based method, Criminisi algorithm is a widely used algorithm. However, the inaccurate completion order and the inefficiency in searching matching patches are two main limitations of Criminisi algorithm. To overcome these shortcomings, in this paper, an exemplar image completion based on evolutionary algorithm is proposed. In the non-textured-based method, total variation method is a typical algorithm. An improved total variation algorithm is proposed in this paper. In the improved algorithm, the diffusion coefficients are defined according to the distance and direction between the damaged pixel and its neighborhood pixel. Experimental results show that the proposed algorithms have better general performance in image completion. And these two new algorithms could improve the experience of network surfing and reduce the network communication cost.     

Scientific workflows for process mining: building blocks,scenarios, and implementation

Over the past decade process mining has emerged as a new analytical discipline able to answer a variety of questions based on event data. Event logs have a very particular structure; events have timestamps, refer to activities and resources, and need to be correlated to form process instances. Process mining results tend to be very different from classical data mining results, e.g., process discovery may yield end-to-end process models capturing different perspectives rather than decision trees or frequent patterns. A process-mining tool like ProM provides hundreds of different process mining techniques ranging from discovery and conformance checking to filtering and prediction. Typically, a combination of techniques is needed and, for every step, there are different techniques that may be very sensitive to parameter settings. Moreover, event logs may be huge and may need to be decomposed and distributed for analysis. These aspects make it very cumbersome to analyze event logs manually. Process mining should be repeatable and automated. Therefore, we propose a framework to support the analysis of process mining workflows. Existing scientific workflow systems and data mining tools are not tailored towards process mining and the artifacts used for analysis (process models and event logs). This paper structures the basic building blocks needed for process mining and describes various analysis scenarios. Based on these requirements we implemented RapidProM, a tool supporting scientific workflows for process mining. Examples illustrating the different scenarios are provided to show the feasibility of the approach.     

A Nom historical document recognition system for digital archiving

A Nom historical document recognition system is being developed for digital archiving that uses image binarization, character segmentation, and character recognition. It incorporates two versions of off-line character recognition: one for automatic recognition of scanned and segmented character patterns (7660 categories) and the other for user handwritten input (32,695 categories). This separation is used since including less frequently appearing categories in automatic recognition increases the misrecognition rate without reliable statistics on the Nom language. Moreover, a user must be able to check the results and identify the correct categories from an extended set of categories, and a user can input characters by hand. Both versions use the same recognition method, but they are trained using different sets of training patterns. Recursive X–Y cut and Voronoi diagrams are used for segmentation; k–d tree and generalized learning vector quantization are used for coarse classification; and the modified quadratic discriminant function is used for fine classification. The system provides an interface through which a user can check the results, change binarization methods, rectify segmentation, and input correct character categories by hand. Evaluation done using a limited number of Nom historical documents after providing ground truths for them showed that the two stages of recognition along with user checking and correction improved the recognition results significantly.     

Textual image compression at low bit rates based on region-of-interest coding

In this paper, we deal with those applications of textual image compression where high compression ratio and maintaining or improving the visual quality and readability of the compressed images are of main concern. In textual images, most of the information exists in the edge regions; therefore, the compression problem can be studied in the framework of region-of-interest (ROI) coding. In this paper, the Set Partitioning in Hierarchical Trees (SPIHT) coder is used in the framework of ROI coding along with some image enhancement techniques in order to remove the leakage effect which occurs in the wavelet-based low-bit-rate compression. We evaluated the compression performance of the proposed method with respect to some qualitative and quantitative measures. The qualitative measures include the averaged mean opinion scores (MOS) curve along with demonstrating some outputs in different conditions. The quantitative measures include two proposed modified PSNR measures and the conventional one. Comparing the results of the proposed method with those of three conventional approaches, DjVu, JPEG2000, and SPIHT coding, showed that the proposed compression method considerably outperformed the others especially from the qualitative aspect. The proposed method improved the MOS by 20 and 30 %, in average, for high- and low-contrast textual images, respectively. In terms of the modified and conventional PSNR measures, the proposed method outperformed DjVu and JPEG2000 up to 0.4 dB for high-contrast textual images at low bit rates. In addition, compressing the high contrast images using the proposed ROI technique, compared to without using this technique, improved the average textual PSNR measure up to 0.5 dB, at low bit rates.     

Maximal limited similarity-based rough set model

Non-symmetric similarity relation-based rough set model (NS-RSM) is viewed as mathematical tool to deal with the analysis of imprecise and uncertain information in incomplete information systems with “?” values. NS-RSM relies on the concept of non-symmetric similarity relation to group equivalent objects and generate knowledge granules that are then used to approximate the target set. However, NS-RSM results in unpromising approximation space when addressing inconsistent data sets that have lots of boundary objects. This is because objects in the same similarity classes are not necessarily similar to each other and may belong to different target classes. To enhance NS-RSM capability, we introduce the maximal limited similarity-based rough set model (MLS-RSM) which describes the maximal collection of indistinguishable objects that are limited tolerance to each other in similarity classes. This allows accurate computation to be done for the approximation space. Furthermore, approximation accuracy comparisons have been conducted among NS-RSM and MLS-RSM. The results demonstrate that MLS-RSM model outperforms NS-RSM and can approximate the target set more efficiently.