Current Chinese Computer Science - Current Issue

Background: In this paper, we present a theoretical discussion on neuromorphic computing circuit dynamic and its relations with AI deep learning neural networks. The hardware implememtations of neuromorphic computing and AI deep learning neuronal networks are discussed.

Objective: The investigation is motivated by the design of a feasible fast and energy-efficient circuit device as well as an efficient training computation method to solve complex classification problems using AI neuronal networks.

Methods: We focus on the investigations of solving pattern classification and recognition problems in real applications from the perspectives of both logic computation view and physical circuit device views. FPGA approaches are considered and a mapping from logic level to physical level is proposed.

Results: A pragmatic mapping method is derived. FPGA method is proposed.

Conclusion: Thus we propose in this paper an approach to solve complex classification problems. First, the neuromorphic computing as a new research area is introduced, including physical circuit properties, memristive device physical properties and the circuit dynamics described by the temporal and spatial (Maxwell) differential equations. Secondly, we show that by using AI deep learning neural networks to train AI neural networks we are able to derive the optimal AI neuron network weights. Last but not the least, we brief a mapping method and show in general how the neuromorphic circuit will work in practice after mapping the weights from AI deep learning neural networks into the neuromorphic circuit synapses/memristors. We also devote our discussions to the physical device feasibility and related matters. The method proposed in this paper is pragmatic and constructive.

Background: Owing to Linguistic Neutrosophic Numbers (LNNs) depicted independently by the truth, indeterminacy, and falsity linguistic variables, they fit in with human thinking and expressing habits to judgments of complicated objects, such as medical diagnosis and Medical Treatment Options (MTOs) for patients in clinical medicine. Unfortunately, existing linguistic neutrosophic Decision Making (DM) approaches have not been applied in medical DM problems so far.

Objective: Then, the LNN multicriteria group DM method especially suits medical DM problems with LNN information since medical DM problems commonly imply inconsistent, incomplete, and indeterminate linguistic information due to the medical DM complexity.

Methods: Therefore, this study proposes two correlation coefficients of linguistic neutrosophic sets (LNSs) and their multicriteria group DM method to deal with the DM problem of MTOs as a new complementary in linguistic medical DM problems. Then, an actual example regarding the DM problem of MTOs is provided to illustrate the applicability of the developed group DM method.

Result: By comparative analysis with existing relative methods in LNN setting, the best MTO for the patient with verruca plantaris is feasible.

Conclusion: The developed DM method is effective in the DM problem of MTOs with LNN information and provides a new way for linguistic medical DM problems.

Background: One of the great challenges in social network analysis is community detection. Community is a group of users which have high intra connections and sparse inter connections. Community detection or clustering reveals community structure of social networks and hidden relationships among their constituents. Nowadays, many different methods are proposed to detect community structures in social networks from different perspectives, but none of them can be a constant winner. Therefore, an ensemble of different methods can potentially improve the final result.

Methods: In this paper, we present a framework for different methods to be combined for community detection. This method is a combination of genetic algorithms, particle swarm optimization, k-means clustering and Louvain clustering algorithms. Our method uses topological and demographic information to identify communities and can automatically determine the number of optimal communities.

Results and Conclusion: Quantitative evaluations based on extensive experiments on Ego-Facebook social network dataset reveals that the method presented in this study achieves favorable results, which are quite superior to other relevant algorithms in the literature.

•   Discovering relationships between individuals by analyzing social networks.

•   Providing identifying communities algorithms based on different clustering methods.

•    An ensemble of community detection consisting of GA, PSO, k-means and Louvain clustering.

•   The proposed method is better than the TSA method at silhouette and modularity criterion.

Demographic information also relates to the profile of users and their shared tweets.

Introduction: In order to ensure the efficiency and cost of dangerous goods warehouse under the premise of safety, this study takes the dangerous goods warehouse as the research object and implements multitask for dangerous goods warehouse with two forklifts.

Methods: This study takes traversal calculation, novel safety calculation formula and scheduling scheme evaluation model as tools to research the forklift scheduling scheme of one-storey packed dangerous goods warehouse.

Results: Optimal scheme and allocation decision model are obtained through numerical simulation. The innovation of this study is giving the safety formula of forklift operation in dangerous goods warehouse and using numerical simulation to obtain the global optimal solution. Furthermore, this study draws on the concept of the travel chain to propose the idea of warehousing chain. At the same time, optimal schemes for multiple dangerous goods inbound and outbound the warehouse are studied.

Background: A Fuzzy Cubic Set (FCS) is composed of a Fuzzy Set (FS) (certain fuzzy numbers) and an Interval-Valued Fuzzy Set (IVFS) (uncertain fuzzy numbers) to describe the hybrid information of both. To enhance the credibility of both, they should be closely related to the measures/degrees of credibility owing to the vagueness and uncertainty of humans’ cognitions regarding the real world.

Objective: This paper presents the notions of a Fuzzy Cubic Credibility Set (FCCS) and a Fuzzy Cubic Credibility Number (FCCN) as the new generalization of the FCS notion to enhance the credibility level of FCS by means of the credibility degrees of both FS and IVFS. Next, we define operations of FCCNs, an expected value of FCCN, and the FCCN Weighted Arithmetic Averaging (FCCNWAA) and FCCN Weighted Geometric Averaging (FCCNWGA) operators for Decision Making (DM) strategy.

Methods: A DM strategy using the FCCNWAA or FCCNWGA operator is proposed to solve multicriteria DM problems in the environment of FCCNs. Then, the proposed DM strategy is applied to a DM example of slope design schemes for an open-pit mine in the environment of FCCNs to reflect the feasibility of the proposed DM strategy.

Results: By comparison with the fuzzy cubic DM strategy, the DM results with and without the degrees of credibility, can impact on the ranking of alternatives in the DM example to reflect the effectiveness of the proposed DM strategy.

Conclusion: However, the highlighting advantage of this study is that the proposed DM strategy not only indicates the degrees of credibility regarding the assessed values of FCNs in the DM process but also enhances the DM reliability in the environment of FCCNs. Hence, the proposed DM strategy is superior to the fuzzy cubic DM strategy in the environment of FCCNs.

Background: More bona fide adjustments aimed at appraising counterparty risks and financial expenses related to over-the-counter derivative have become indispensable after the European sovereign debt catastrophe and the 2007/08’s worldwide fiscal crisis. The most notable measures include DVA, CVA, and FVA.

Methods: This paper advocates the application of the XVA scheme to assess CVA, DVA, and FVA for managing risk and pricing of financial or OTC derivatives.

Results and Discussion: A foundation formula is formulated and tested against different risk scenarios of CVA, DVA, FVA, and KVA using cross-referenced data. Practical pieces of advice are provided for the real industry application of XVA.

Conclusion: Compared to traditional risk management in the financial market where funding risk, credit risk, and default risk are accounted separately, the approach proposed by the current study monitors the multiple types of risk in a comprehensive framework and is more practically effective from a financial operation point of view.

Background: In the era of information overload it is very difficult for a human reader to make sense of the vast information available on the internet quickly. Even for a specific domain like a college or university website, it may be difficult for a user to browse through all the links to quickly get the relevant answers.

Objective: In this scenario, the design of a chat-bot which can answer questions related to college information and compare between colleges will be very useful and novel.

Methods: In this paper, a novel conversational interface chat-bot application with information retrieval and text summarization skill is designed and implemented. Firstly, this chat-bot has a simple dialog skill; when it can understand the user query intent, it responds from the stored collection of answers. Secondly, for unknown queries, this chat-bot can search the internet, and then perform text summarization using advanced techniques of natural language processing (NLP) and text mining (TM).

Results: The advancement of NLP capability of information retrieval and text summarization using machine learning techniques of Latent Semantic Analysis (LSI), Latent Dirichlet Allocation (LDA), Word2Vec, Global Vector (GloVe) and TextRank is reviewed and compared in this paper first before implementing them for the chat-bot design. This chat-bot improves user experience tremendously by getting answers to specific queries concisely which takes less time than to read the entire document. Students, parents and faculty can get the answers for a variety of information like admission criteria, fees, course offerings, notice board, attendance, grades, placements, faculty profile, research papers, patents, etc. more efficiently.

Conclusion: The purpose of this paper was to follow the advancement in NLP technologies and implement them in a novel application.

Background: Owing to the ability of a deep learning algorithm to identify objects and the related detection technology of security inspection equipment, in this paper, we propose a progressive object recognition method that considers local information of objects.

Methods: First, we construct an X-Base model by cascading multiple convolutions and pooling layers to obtain the feature mapping image. Moreover, we provide a “segmented convolution, unified recognition” strategy to detect the size of the objects.

Results: Experimental results show that this method can effectively identify the specifications of bags passing through the security inspection equipment. Compared with the traditional VGG and progressive VGG recognition methods, the proposed method achieves advantages in terms of efficiency and concurrency.

Conclusion: This study provides a method to gradually recognize objects and can potentially assist the operators in identifying prohibited objects.

Background: Crop diseases are a primary hazard to nutrient safety, which proves to be a serious problem in many places in the world due to the unavailability of essential aid. Typically agriculturalists or specialists perceive the plants with a naked eye for detection and identification of an illness. Machine vision models, in specific Convolutional Neural Networks (CNNs) have directed an impact in feature extraction to a greater extent. Since 2015, numerous solicitations for the automatic classification and recognition of crop illnesses have been established.

Methods: In this paper, we proposed, analyzed, and assessed various state-of-the-art models proposed over a decade. These models are pre-trained with the finest parameters where we modeled a design-oriented method with numerous leaf-images and classified them into infection and healthy class for each type of leaf independently.

Results: Through our examination, we concluded that VGG models stand-alone with many cited prototypes and give on par results. As declared, these VGG models (VGG16 and VGG19) are utilized for feature extraction, and further, we augmented a set of dense layers and train them consequently for classification. The performances of various machine vision prototypes were pictorially perceived and their sophisticated architecture is not only capable of extracting detailed features but also repressed many loop-holes. The performance is assessed and computed for several types of leaf images and the accuracy scores attained were more than 97.5% for VGG16 and 96.72% for VGG19.

Conclusion: AUC-ROC curves were portrayed to illustrate its inspiration in defining an accurate classification where VGG16 and VGG19 have at least 96.6% and 95% area under the curve (AUC) which resembles their robustness.

Objectives: The purpose of this study is to analyze the performance of a one-megawatt photovoltaic power plant in Arak-Iran, according to IEC-61724 standard, using data recorded over a year. The photovoltaic plant of Arak is located at coordinates 34.0954° N and 49.7013° E. This power plant is the first-megawatt photovoltaic power plant in Iran which two types of modules are used and it was constructed by the New Energy Agency and the Power Research Center under the supervision of the Ministry of Energy in 2016. In this plant, a combination of monocrystal and polycrystalline modules is used, and the annual output is 1756 MWh.

Methods: The combination of modules is based on the 1920 modules of 250 W of polycrystalline and 260 modules of 260 W of monocrystal in the construction of the power plant. There are also 4 inverters and a 1250 KVA dry power trans-former. The plant has suitable productivity, with a performance ratio equal to 0.8 and a final yield of 4.57.

Results and Conclusion: Ultimately the PV plant is simulated by PVsyst and the results are compared with monitored records which indi-cated the appropriate accuracy of the collected data. The calculated performance ratio for the power plant by PVsyst is 81.2% and has a 1.5% difference with the monitored totals. The energy supplied during one year by the power plant is 1756 MWh, whereas the prediction of annual energy yield that entered to the grid is equal to 1757 MWh.