Search results

The trend of “Deep Learning for Internet of Things (IoT)” has gained fresh momentum with enormous upcoming applications employing these models as their processing engine and Cloud as their resource giant. But this picture leads to underutilization of ever-increasing device pool of IoT that has already passed 15 billion mark in 2015. Thus, it is high time to explore a different approach to tackle this issue, keeping in view the characteristics and needs of the two fields. Processing at the Edge can boost applications with real-time deadlines while complementing security.

This review paper contributes towards three cardinal directions of research in the field of DL for IoT. The first section covers the categories of IoT devices and how Fog can aid in overcoming the underutilization of millions of devices, forming the realm of the things for IoT. The second direction handles the issue of immense computational requirements of DL models by uncovering specific compression techniques. An appropriate combination of these techniques, including regularization, quantization, and pruning, can aid in building an effective compression pipeline for establishing DL models for IoT use-cases. The third direction incorporates both these views and introduces a novel approach of parallelization for setting up a distributed systems view of DL for IoT.

DL models are growing deeper with every passing year. Well-coordinated distributed execution of such models using Fog displays a promising future for the IoT application realm. It is realized that a vertically partitioned compressed deep model can handle the trade-off between size, accuracy, communication overhead, bandwidth utilization, and latency but at the expense of an additionally considerable memory footprint. To reduce the memory budget, we propose to exploit Hashed Nets as potentially favorable candidates for distributed frameworks. However, the critical point between accuracy and size for such models needs further investigation.

To the best of our knowledge, no study has explored the inherent parallelism in deep neural network architectures for their efficient distribution over the Edge-Fog continuum. Besides covering techniques and frameworks that have tried to bring inference to the Edge, the review uncovers significant issues and possible future directions for endorsing deep models as processing engines for real-time IoT. The study is directed to both researchers and industrialists to take on various applications to the Edge for better user experience.

The problem of image retrieval and image description exists in various fields. In this paper, a model of content-based image retrieval and image content extraction based on the KD-Tree structure was proposed.

A Random Forest structure was built to classify the objects on each image on the basis of the balanced multibranch KD-Tree structure. From that purpose, a KD-Tree structure was generated by the Random Forest to retrieve a set of similar images for an input image. A KD-Tree structure is applied to determine a relationship word at leaves to extract the relationship between objects on an input image. An input image content is described based on class names and relationships between objects.

A model of image retrieval and image content extraction was proposed based on the proposed theoretical basis; simultaneously, the experiment was built on multi-object image datasets including Microsoft COCO and Flickr with an average image retrieval precision of 0.9028 and 0.9163, respectively. The experimental results were compared with those of other works on the same image dataset to demonstrate the effectiveness of the proposed method.

A balanced multibranch KD-Tree structure was built to apply to relationship classification on the basis of the original KD-Tree structure. Then, KD-Tree Random Forest was built to improve the classifier performance and retrieve a set of similar images for an input image. Concurrently, the image content was described in the process of combining class names and relationships between objects.

Traditionally, loan officers use different credit scoring models to complement judgmental methods to classify consumer loan applications. This study explores the use of decision trees, AdaBoost, and support vector machines (SVMs) to identify potential bad loans. Our results show that AdaBoost does provide an improvement over simple decision trees as well as SVM models in predicting good credit clients and bad credit clients. To cross-validate our results, we use k-fold classification methodology.

Inverse problems deal with determining the causes on the basis of knowing their effects. The object of the inverse parameter estimation problem is to fix the thermal material parameters (the cause) on the strength of a given observation of the temperature history at one or more interior points (the effect). This paper demonstrates two novel approaches to the inverse problems. These approaches use two artificial intelligence mechanisms: neural network and genetic algorithm. Examples shown in this paper give a comparison of results obtained by both of these methods. The numerical technique of neural networks evolved from the effort to model the function of the human brain and the genetic algorithms model the evolutional process of nature. Both of the presented approaches can lead to a solution without having problems with the stability of the inverse task. Both methods are suitable for parallel processing and are advantageous for a multiprocessor computer architecture.

This study aims to develop a push–pull factors theory of women entrepreneurship, to identify and prioritize the factors influencing Emirati women entrepreneurs, and also aims to implement the proposed theory in two cases: Emirati women entrepreneurs with business family and non-business family backgrounds.

The analytic hierarchy process (AHP) model was developed with 6 criteria and 19 sub-criteria, based upon the findings of previous studies. Data were collected using a questionnaire survey given to 20 Emirati women entrepreneurs in the United Arab Emirates (UAE). The respondents were selected on the basis of their family backgrounds. The data collected were interpreted and a priority vector was assigned to the criteria and sub-criteria.

A well-researched methodology was used for the synthesis of priorities and the measurement of consistencies. The findings show that education, skills and training are the three main criteria considered to be the most important factors that influence the growth and success of Emirati women entrepreneurs.

The model can be used by authors for future academic and entrepreneurial studies. The findings interpreted can help policymakers and related associations develop various policies based on the specific factors found to empower Emirati women entrepreneurs in an effective manner. This process will increase the participation of Emirati women in the entrepreneurial field. The research model had limited dimensions and the findings cannot be generalized. Hence, it would be valuable to conduct future study in other countries to generalize the findings. The model can be enhanced by including other factors, and alternatives could be based on types of sectors.

This study is the first of its kind to present an AHP model that contains most dimensions influencing the success and growth of Emirati women entrepreneurs and prioritizes the dimensions based on their importance.

This paper aims to explore the effectiveness of an artificial neural network (ANN) in predicting fraudulent financial reporting in small market capitalization companies in Malaysia.

Based on the concepts of ANN, a mathematical model was developed to compare non-fraud and fraud companies selected from among small market capitalization companies in Malaysia; the fraud companies had already been charged by the Securities Commission for falsification of financial statements. Ten financial ratios are used as fraud risk indicators to predict fraudulent financial reporting using ANN.

The findings indicate that the proposed ANN methodology outperforms other statistical techniques widely used for predicting fraudulent financial reporting.

The study is one of few to adopt the ANN approach for the prediction of financial reporting fraud.

This paper aims to propose a classification method for steel strip surface defects based on a mixed attention mechanism to achieve fast and accurate classification performance. The traditional method of classifying surface defects of hot-rolled steel strips has the problems of low recognition accuracy and low efficiency in the industrial complex production environment.

The authors selected min–max scaling comparison method to filter the training results of multiple network models on the steel strip surface defect data set. Then, the best comprehensive performance model EfficientNet-B0 was refined. Based on this, the authors proposed two mixed attention addition methods, which include squeeze-excitation spatial mixed module and multilayer mixed attention mechanism (MMAM) module, respectively.

With these two methods, the authors achieved 96.72% and 97.70% recognition accuracy on the steel strip data set after data augmentation for adapting to the complex production environment, respectively. Using the transfer learning method, the EfficientNet-B0 based on MMAM obtained 100% recognition accuracy.

This study not only focuses on improving the recognition accuracy of the network model itself but also considers other performance indicators of the network, which are rarely considered by many researchers. The authors further improve the intelligent production technique and address this issue. Both methods proposed in this paper can be applied to embedded equipment, which can effectively improve steel strip factory production efficiency and reduce material and time loss.

This paper aims to solve the problem between detection efficiency and performance in grasp commodities rapidly. A fast detection and grasping method based on improved faster R-CNN is purposed and applied to the mobile manipulator to grab commodities on the shelf.

To reduce the time cost of algorithm, a new structure of neural network based on faster R CNN is designed. To select the anchor box reasonably according to the data set, the data set-adaptive algorithm for choosing anchor box is presented; multiple models of ten types of daily objects are trained for the validation of the improved faster R-CNN. The proposed algorithm is deployed to the self-developed mobile manipulator, and three experiments are designed to evaluate the proposed method.

The result indicates that the proposed method is successfully performed on the mobile manipulator; it not only accomplishes the detection effectively but also grasps the objects on the shelf successfully.

The proposed method can improve the efficiency of faster R-CNN, maintain excellent performance, meet the requirement of real-time detection, and the self-developed mobile manipulator can accomplish the task of grasping objects.

Fitness service companies often promote the companies' personal training service by attributing trainers' competent characteristics, qualifications or/and service provision to their effort or talent. This promotion is called performance attribution promotion. Utilizing attribution theory and the theory's adjacent studies, this study investigated whether and why performance attribution promotion affects consumers' service purchase of personal fitness training.

The authors developed the experimental stimuli of performance attribution promotion and validated those through a pretest (N = 200). Using the validated stimuli, the authors conducted an experiment with employing a single factor between-subject design (performance attribution promotion: effort vs talent) based on random assignment (N = 200).

The analysis results revealed that attributing trainers' competent characteristics, qualifications or/and service provision to effort (vs talent) leads to a higher level of service registration intention. Moreover, this effect was mediated by the perceived teaching expertise but not by the perceived teaching trustworthiness.

These findings enrich the literature by illuminating a new mechanism and consequence of performance attribution promotion. The authors' study also extends the marketing studies related to expertise perception by presenting the attribution of visible features as one of the characteristics determining expertise perception. Finally, the authors' findings also have implications for fitness service companies and other stakeholders that seek to effectively leverage trainers' competent outcomes for consumer acquisition.