Search results

Robot vision is a fundamental device for human–robot interaction and robot complex tasks. In this paper, the authors aim to use Kinect and propose a feature graph fusion (FGF) for robot recognition.

The feature fusion utilizes red green blue (RGB) and depth information to construct fused feature from Kinect. FGF involves multi-Jaccard similarity to compute a robust graph and word embedding method to enhance the recognition results.

The authors also collect DUT RGB-Depth (RGB-D) face data set and a benchmark data set to evaluate the effectiveness and efficiency of this method. The experimental results illustrate that FGF is robust and effective to face and object data sets in robot applications.

The authors first utilize Jaccard similarity to construct a graph of RGB and depth images, which indicates the similarity of pair-wise images. Then, fusion feature of RGB and depth images can be computed by the Extended Jaccard Graph using word embedding method. The FGF can get better performance and efficiency in RGB-D sensor for robots.

The purpose of this paper is to collect and understand the nature of real cases of author name variants that have often appeared in bibliographic digital libraries (DLs) as a case study of the name authority control problem in DLs.

To find a sample of name variants across DLs (e.g. DBLP and ACM) and in a single DL (e.g. ACM), the approach is based on two bipartite matching algorithms: Maximum Weighted Bipartite Matching and Maximum Cardinality Bipartite Matching.

First, the authors validated the effectiveness and efficiency of the bipartite matching algorithms. The authors also studied the nature of real cases of author name variants that had been found across DLs (e.g. ACM, CiteSeer and DBLP) and in a single DL.

To the best of the authors knowledge, there is less research effort to understand the nature of author name variants shown in DLs. A thorough analysis can help focus research effort on real problems that arise when the authors perform duplicate detection methods.

The design of a cellular manufacturing system requires that a machine population be partitioned into machine groups called manufacturing cells. A new graph partitioning heuristic is proposed to solve the manufacturing cell formation problem (MCFP). In the proposed heuristic, The MCFP is represented by a graph whose node set represents the machine cluster and edge set represents the machine‐pair association weights. A graph partitioning approach is used to form the manufacturing cells. This approach offers improved design flexibility by allowing a variety of design parameters to be controlled during cell formation. The effectiveness of the heuristic is demonstrated by comparing it to two MCFP published solution methods using several problems from the literature.

Multi-view fuzzy clustering algorithms are not widely used in image segmentation, and many of these algorithms are lacking in robustness. The purpose of this paper is to investigate a new algorithm that can segment the image better and retain as much detailed information about the image as possible when segmenting noisy images.

The authors present a novel multi-view fuzzy c-means (FCM) clustering algorithm that includes an automatic view-weight learning mechanism. Firstly, this algorithm introduces a view-weight factor that can automatically adjust the weight of different views, thereby allowing each view to obtain the best possible weight. Secondly, the algorithm incorporates a weighted fuzzy factor, which serves to obtain local spatial information and local grayscale information to preserve image details as much as possible. Finally, in order to weaken the effects of noise and outliers in image segmentation, this algorithm employs the kernel distance measure instead of the Euclidean distance.

The authors added different kinds of noise to images and conducted a large number of experimental tests. The results show that the proposed algorithm performs better and is more accurate than previous multi-view fuzzy clustering algorithms in solving the problem of noisy image segmentation.

Most of the existing multi-view clustering algorithms are for multi-view datasets, and the multi-view fuzzy clustering algorithms are unable to eliminate noise points and outliers when dealing with noisy images. The algorithm proposed in this paper has stronger noise immunity and can better preserve the details of the original image.

The COVID-19 pandemic has sparked a remarkable volume of research literature, and scientists are increasingly in need of intelligent tools to cut through the noise and uncover relevant research directions. As a response, the authors propose a novel framework. In this framework, the authors develop a novel weighted semantic graph model to compress the research studies efficiently. Also, the authors present two analyses on this graph to propose alternative ways to uncover additional aspects of COVID-19 research.

The authors construct the semantic graph using state-of-the-art natural language processing (NLP) techniques on COVID-19 publication texts (>100,000 texts). Next, the authors conduct an evolutionary analysis to capture the changes in COVID-19 research across time. Finally, the authors apply a link prediction study to detect novel COVID-19 research directions that are so far undiscovered.

Findings reveal the success of the semantic graph in capturing scientific knowledge and its evolution. Meanwhile, the prediction experiments provide 79% accuracy on returning intelligible links, showing the reliability of the methods for predicting novel connections that could help scientists discover potential new directions.

To the authors’ knowledge, this is the first study to propose a holistic framework that includes encoding the scientific knowledge in a semantic graph, demonstrates an evolutionary examination of past and ongoing research and offers scientists with tools to generate new hypotheses and research directions through predictive modeling and deep machine learning techniques.

What makes investors tick? Largely counter-intuitive compared to the findings of most past research, this study explores the possibility that funding investors invest in companies based on social relationships, which could be positive or negative, similar or dissimilar. The purpose of this paper is to build a social network graph using data from CrunchBase, the largest public database with profiles about companies. The authors combine social network analysis with the study of investing behavior in order to explore how similarity between investors and companies affects investing behavior through social network analysis.

This study crawls and analyzes data from CrunchBase and builds a social network graph which includes people, companies, social links and funding investment links. The problem is then formalized as a link (or relationship) prediction task in a social network to model and predict (across various machine learning methods and evaluation metrics) whether an investor will create a link to a company in the social network. Various link prediction techniques such as common neighbors, shortest path, Jaccard Coefficient and others are integrated to provide a holistic view of a social network and provide useful insights as to how a pair of nodes may be related (i.e., whether the investor will invest in the particular company at a time) within the social network.

This study finds that funding investors are more likely to invest in a particular company if they have a stronger social relationship in terms of closeness, be it direct or indirect. At the same time, if investors and companies share too many common neighbors, investors are less likely to invest in such companies.

The author’s study is among the first to use data from the largest public company profile database of CrunchBase as a social network for research purposes. The author ' s also identify certain social relationship factors that can help prescribe the investor funding behavior. Authors prediction strategy based on these factors and modeling it as a link prediction problem generally works well across the most prominent learning algorithms and perform well in terms of aggregate performance as well as individual industries. In other words, this study would like to encourage companies to focus on social relationship factors in addition to other factors when seeking external funding investments.

The purpose of this study is to develop a hybrid algorithm for segmenting tumor from ultrasound images of the liver.

After collecting the ultrasound images, contrast-limited adaptive histogram equalization approach (CLAHE) is applied as preprocessing, in order to enhance the visual quality of the images that helps in better segmentation. Then, adaptively regularized kernel-based fuzzy C means (ARKFCM) is used to segment tumor from the enhanced image along with local ternary pattern combined with selective level set approaches.

The proposed segmentation algorithm precisely segments the tumor portions from the enhanced images with lower computation cost. The proposed segmentation algorithm is compared with the existing algorithms and ground truth values in terms of Jaccard coefficient, dice coefficient, precision, Matthews correlation coefficient, f-score and accuracy. The experimental analysis shows that the proposed algorithm achieved 99.18% of accuracy and 92.17% of f-score value, which is better than the existing algorithms.

From the experimental analysis, the proposed ARKFCM with enhanced level set algorithm obtained better performance in ultrasound liver tumor segmentation related to graph-based algorithm. However, the proposed algorithm showed 3.11% improvement in dice coefficient compared to graph-based algorithm.

The image preprocessing is carried out using CLAHE algorithm. The preprocessed image is segmented by employing selective level set model and Local Ternary Pattern in ARKFCM algorithm. In this research, the proposed algorithm has advantages such as independence of clustering parameters, robustness in preserving the image details and optimal in finding the threshold value that effectively reduces the computational cost.

The innovation in big data is increasing day by day in such a way that the conventional software tools face several problems in managing the big data. Moreover, the occurrence of the imbalance data in the massive data sets is a major constraint to the research industry.

The purpose of the paper is to introduce a big data classification technique using the MapReduce framework based on an optimization algorithm. The big data classification is enabled using the MapReduce framework, which utilizes the proposed optimization algorithm, named chicken-based bacterial foraging (CBF) algorithm. The proposed algorithm is generated by integrating the bacterial foraging optimization (BFO) algorithm with the cat swarm optimization (CSO) algorithm. The proposed model executes the process in two stages, namely, training and testing phases. In the training phase, the big data that is produced from different distributed sources is subjected to parallel processing using the mappers in the mapper phase, which perform the preprocessing and feature selection based on the proposed CBF algorithm. The preprocessing step eliminates the redundant and inconsistent data, whereas the feature section step is done on the preprocessed data for extracting the significant features from the data, to provide improved classification accuracy. The selected features are fed into the reducer for data classification using the deep belief network (DBN) classifier, which is trained using the proposed CBF algorithm such that the data are classified into various classes, and finally, at the end of the training process, the individual reducers present the trained models. Thus, the incremental data are handled effectively based on the training model in the training phase. In the testing phase, the incremental data are taken and split into different subsets and fed into the different mappers for the classification. Each mapper contains a trained model which is obtained from the training phase. The trained model is utilized for classifying the incremental data. After classification, the output obtained from each mapper is fused and fed into the reducer for the classification.

The maximum accuracy and Jaccard coefficient are obtained using the epileptic seizure recognition database. The proposed CBF-DBN produces a maximal accuracy value of 91.129%, whereas the accuracy values of the existing neural network (NN), DBN, naive Bayes classifier-term frequency–inverse document frequency (NBC-TFIDF) are 82.894%, 86.184% and 86.512%, respectively. The Jaccard coefficient of the proposed CBF-DBN produces a maximal Jaccard coefficient value of 88.928%, whereas the Jaccard coefficient values of the existing NN, DBN, NBC-TFIDF are 75.891%, 79.850% and 81.103%, respectively.

In this paper, a big data classification method is proposed for categorizing massive data sets for meeting the constraints of huge data. The big data classification is performed on the MapReduce framework based on training and testing phases in such a way that the data are handled in parallel at the same time. In the training phase, the big data is obtained and partitioned into different subsets of data and fed into the mapper. In the mapper, the features extraction step is performed for extracting the significant features. The obtained features are subjected to the reducers for classifying the data using the obtained features. The DBN classifier is utilized for the classification wherein the DBN is trained using the proposed CBF algorithm. The trained model is obtained as an output after the classification. In the testing phase, the incremental data are considered for the classification. New data are first split into subsets and fed into the mapper for classification. The trained models obtained from the training phase are used for the classification. The classified results from each mapper are fused and fed into the reducer for the classification of big data.

Because of short life cycle and fluctuating greatly in total sales volumes (TSV), it is difficult to accumulate enough sales data and mine an attribute set reflecting the common needs of all consumers for a kind of experiential product with short life cycle (EPSLC). Methods for predicting TSV of long-life-cycle products may not be suitable for EPSLC. Furthermore, point prediction cannot obtain satisfactory prediction results because information available before production is inadequate. Thus, this paper aims at proposing and verifying a novel interval prediction method (IPM).

Because interval prediction may satisfy requirements of preproduction investment decision-making, interval prediction was adopted, and then the prediction difficult was converted into a classification problem. The classification was designed by comparing similarities in attribute relationship patterns between a new EPSLC and existing product groups. The product introduction may be written or obtained before production and thus was designed as primary source information. IPM was verified by using data of crime movies released in China from 2013 to 2017.

The IPM is valid, which uses product introduction as input, classifies existing products into three groups with different TSV intervals, mines attribute relationship patterns using content and association analyses and compares similarities in attribute relationship patterns – to predict TSV interval of a new EPSLC before production.

Different from other studies, the IPM uses product introduction to mine attribute relationship patterns and compares similarities in attribute relationship patterns to predict the interval values. It has a strong applicability in data content and structure and may realize rolling prediction.

Massive amounts of data, manifesting in various forms, are being produced on the Web every minute and becoming the new standard. Exploring these information sources distributed in different Web segments in a unified way is becoming a core task for a variety of users’ and companies’ scenarios. However, knowledge creation and exploration from distributed Web data sources is a challenging task. Several data integration conflicts need to be resolved and the knowledge needs to be visualized in an intuitive manner. The purpose of this paper is to extend the authors’ previous integration works to address semantic knowledge exploration of enterprise data combined with heterogeneous social and linked Web data sources.

The authors synthesize information in the form of a knowledge graph to resolve interoperability conflicts at integration time. They begin by describing KGMap, a mapping model for leveraging knowledge graphs to bridge heterogeneous relational, social and linked web data sources. The mapping model relies on semantic similarity measures to connect the knowledge graph schema with the sources' metadata elements. Then, based on KGMap, this paper proposes KeyFSI, a keyword-based semantic search engine. KeyFSI provides a responsive faceted navigating Web user interface designed to facilitate the exploration and visualization of embedded data behind the knowledge graph. The authors implemented their approach for a business enterprise data exploration scenario where inputs are retrieved on the fly from a local customer relationship management database combined with the DBpedia endpoint and the Facebook Web application programming interface (API).

The authors conducted an empirical study to test the effectiveness of their approach using different similarity measures. The observed results showed better efficiency when using a semantic similarity measure. In addition, a usability evaluation was conducted to compare KeyFSI features with recent knowledge exploration systems. The obtained results demonstrate the added value and usability of the contributed approach.

Most state-of-the-art interfaces allow users to browse one Web segment at a time. The originality of this paper lies in proposing a cost-effective virtual on-demand knowledge creation approach, a method that enables organizations to explore valuable knowledge across multiple Web segments simultaneously. In addition, the responsive components implemented in KeyFSI allow the interface to adequately handle the uncertainty imposed by the nature of Web information, thereby providing a better user experience.