Search results for: Generative Adversarial Network

Authors: Mohammad Al-Fayoumi, Soumya Banerjee, Jr., P. K. Mahanti

Abstract:

A social network is a set of people or organization or other social entities connected by some form of relationships. Analysis of social network broadly elaborates visual and mathematical representation of that relationship. Web can also be considered as a social network. This paper presents an innovative approach to analyze a social network using a variant of existing ant colony optimization algorithm called as Clever Ant Colony Metaphor. Experiments are performed and interesting findings and observations have been inferred based on the proposed model.

Keywords: Social Network, Ant Colony, Maximum Clique, Sub graph, Clever Ant colony.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1985

Authors: Yifan Fan, Xudong Luo, Pingping Lin

Abstract:

An essential task in the field of artificial intelligence is to allow computers to interact with people through natural language. Therefore, researches such as virtual assistants and dialogue systems have received widespread attention from industry and academia. The response generation plays a crucial role in dialogue systems, so to push forward the research on this topic, this paper surveys various methods for response generation. We sort out these methods into three categories. First one includes finite state machine methods, framework methods, and instance methods. The second contains full-text indexing methods, ontology methods, vast knowledge base method, and some other methods. The third covers retrieval methods and generative methods. We also discuss some hybrid methods based knowledge and deep learning. We compare their disadvantages and advantages and point out in which ways these studies can be improved further. Our discussion covers some studies published in leading conferences such as IJCAI and AAAI in recent years.

Keywords: Retrieval, generative, deep learning, response generation, knowledge.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1204

Authors: B. S. Yew , B. L. Ong , R. B. Ahmad

Abstract:

This paper describes the performance of TCP Vegas over the wireless IPv6 network. The performance of TCP Vegas is evaluated using network simulator (ns-2). The simulation experiment investigates how packet spacing affects the network delay, network throughput and network efficiency of TCP Vegas. Moreover, we investigate how the variable FTP packet sizes affect the network performance. The result of the simulation experiment shows that as the packet spacing is implements, the network delay is reduces, network throughput and network efficiency is optimizes. As the FTP packet sizes increase, the ratio of delay per throughput decreases. From the result of experiment, we propose the appropriate packet size in transmitting file transfer protocol application using TCP Vegas with packet spacing enhancement over wireless IPv6 environment in ns-2. Additionally, we suggest the appropriate ratio in determining the appropriate RTT and buffer size in a network.

Keywords: TCP Vegas, Packet Spacing, Packet Size, Wireless IPv6, ns-2

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1960

Authors: H. Shayeghi, M. Mahdavi, H. Haddadian

Abstract:

Transmission network expansion planning (TNEP) is an important component of power system planning that its task is to minimize the network construction and operational cost while satisfying the demand increasing, imposed technical and economic conditions. Up till now, various methods have been presented to solve the static transmission network expansion planning (STNEP) problem. But in all of these methods, the lines adequacy rate has not been studied after the planning horizon, i.e. when the expanded network misses its adequacy and needs to be expanded again. In this paper, in order to take transmission lines condition after expansion in to account from the line loading view point, the adequacy of transmission network is considered for solution of STNEP problem. To obtain optimal network arrangement, a decimal codification genetic algorithm (DCGA) is being used for minimizing the network construction and operational cost. The effectiveness of the proposed idea is tested on the Garver's six-bus network. The results evaluation reveals that the annual worth of network adequacy has a considerable effect on the network arrangement. In addition, the obtained network, based on the DCGA, has lower investment cost and higher adequacy rate. Thus, the network satisfies the requirements of delivering electric power more safely and reliably to load centers.

Keywords: STNEP Problem, Network Adequacy, DCGA.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1426

Authors: Ahmed Jedidi, Mesfer Mohammed Alshamrani, Alwi Mohammad A. Bamhdi

Abstract:

Optical performance monitoring and optical network management are essential in building a reliable, high-capacity, and service-differentiation enabled all-optical network. One of the serious problems in this network is the fact that optical crosstalk is additive, and thus the aggregate effect of crosstalk over a whole AON may be more nefarious than a single point of crosstalk. As results, we note a huge degradation of the Quality of Service (QoS) in our network. For that, it is necessary to identify and monitor the impairments in whole network. In this way, this paper presents new system to identify and monitor crosstalk in AONs in real-time fashion. particular, it proposes a new technique to manage intra-crosstalk in objective to relax QoS of the network.

Keywords: All-optical networks, optical crosstalk, optical cross-connect, crosstalk, monitoring crosstalk.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1718

Authors: Ei Ei Phyu, Kyaw Myo Lin, Thin Thin Moe

Abstract:

The paper presents an approach to improve the reliability and reduce line losses of practical distribution system applying network reconfiguration. The change of the topology redirects the power flow within the distribution network to obtain better performance of the system. Practical distribution network (Pyigyitagon Industrial Zone (I)) is used as the case study network. The detailed calculations of the reliability indices are done by using analytical method and power flow calculation is performed by Newton-Rephason solver. The comparison of various network reconfiguration techniques are described with respect to power loss and reliability index levels. Finally, the optimal reconfigured network is selected among difference cases based on the two factors: the most reliable network and the least loss minimization.

Keywords: Distribution system reliability, loss reduction, network reconfiguration, reliability enhancement, reliability indices.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 887

Authors: Jong-Whoi Shin, Soon-Tai Park, Chong-Sun Hwang

Abstract:

Active network was developed to solve the problem of the current sharing-based network–difficulty in applying new technology, service or standard, and duplicated operation at several protocol layers. Active network can transport the packet loaded with the executable codes, which enables to change the state of the network node. However, if the network node is placed in the sharing-based network, security and safety issues should be resolved. To satisfy this requirement, various security aspects are required such as authentication, authorization, confidentiality and integrity. Among these security components, the core factor is the encryption key. As a result, this study is designed to propose the scheme that manages the encryption key, which is used to provide security of the comprehensive active directory, based on the domain.

Keywords: Active Network, Domain-based Key Management, Security Components.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1673

Authors: Yong-Quan Zhou, Deng-Xu He, Zheng Nong

Abstract:

In this paper two models using a functional network were employed to solving classification problem. Functional networks are generalized neural networks, which permit the specification of their initial topology using knowledge about the problem at hand. In this case, and after analyzing the available data and their relations, we systematically discuss a numerical analysis method used for functional network, and apply two functional network models to solving XOR problem. The XOR problem that cannot be solved with two-layered neural network can be solved by two-layered functional network, which reveals a potent computational power of functional networks, and the performance of the proposed model was validated using classification problems.

Keywords: Functional network, neural network, XOR problem, classification, numerical analysis method.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1310

Authors: Motonobu Hattori

Abstract:

In neural networks, when new patterns are learned by a network, the new information radically interferes with previously stored patterns. This drawback is called catastrophic forgetting or catastrophic interference. In this paper, we propose a biologically inspired neural network model which overcomes this problem. The proposed model consists of two distinct networks: one is a Hopfield type of chaotic associative memory and the other is a multilayer neural network. We consider that these networks correspond to the hippocampus and the neocortex of the brain, respectively. Information given is firstly stored in the hippocampal network with fast learning algorithm. Then the stored information is recalled by chaotic behavior of each neuron in the hippocampal network. Finally, it is consolidated in the neocortical network by using pseudopatterns. Computer simulation results show that the proposed model has much better ability to avoid catastrophic forgetting in comparison with conventional models.

Keywords: catastrophic forgetting, chaotic neural network, complementary learning systems, dual-network

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2102

Authors: Tet H. Yeap

Abstract:

An analog restricted Hopfield Network is presented in this paper. It consists of two layers of nodes, visible and hidden nodes, connected by directional weighted paths forming a bipartite graph with no intralayer connection. An energy or Lyapunov function was derived to show that the proposed network will converge to stable states. By introducing hidden nodes, the proposed network can be trained to store patterns and has increased memory capacity. Training to be an associative memory, simulation results show that the associative memory performs better than a classical Hopfield network by being able to perform better memory recall when the input is noisy.

Keywords: Associative memory, Hopfield network, Lyapunov function, Restricted Hopfield network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 488

Authors: Payam Nabhani, Sima Radmanesh

Abstract:

Opportunistic network is a kind of Delay Tolerant Networks (DTN) where the nodes in this network come into contact with each other opportunistically and communicate wirelessly and, an end-to-end path between source and destination may have never existed, and disconnection and reconnection is common in the network. In such a network, because of the nature of opportunistic network, perhaps there is no a complete path from source to destination for most of the time and even if there is a path; the path can be very unstable and may change or break quickly. Therefore, routing is one of the main challenges in this environment and, in order to make communication possible in an opportunistic network, the intermediate nodes have to play important role in the opportunistic routing protocols. In this paper we proposed an Adaptive Fuzzy Routing in opportunistic network (AFRON). This protocol is using the simple parameters as input parameters to find the path to the destination node. Using Message Transmission Count, Message Size and Time To Live parameters as input fuzzy to increase delivery ratio and decrease the buffer consumption in the all nodes of network.

Keywords: Opportunistic Routing, Fuzzy Routing, Opportunistic Network, Message Routing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1537

Authors: Ahmad Anzaar, Husain Shahnawaz, Chand Mukesh, S. C. Gupta, R. Gowri, H. L. Mandoria

Abstract:

Due to insufficient frequency band and tremendous growth of the mobile users, complex computation is needed for the use of resources. Long distance communication began with the introduction of telegraphs and simple coded pulses, which were used to transmit short messages. Since then numerous advances have rendered reliable transfer of information both easier and quicker. Wireless network refers to any type of computer network that is wireless, and is commonly associated with a telecommunications network whose interconnections between nodes is implemented without the use of wires. Wireless network can be broadly categorized in infrastructure network and infrastructure less network. Infrastructure network is one in which we have a base station to serve the mobile users and in the infrastructure less network is one in which no infrastructure is available to serve the mobile users this kind of networks are also known as mobile Adhoc networks. In this paper we have simulated the result for different scenarios with protocols like AODV and DSR; we simulated the result for throughput, delay and receiving traffic in the given scenario.

Keywords: Adhoc network, AODV, DSR. mobility.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2109

Authors: Masatora Daito, Hisashi Kojima

Abstract:

In this research, the authors analyze network stability using agent-based simulation. Firstly, the authors focus on analyzing large networks (eight agents) by connecting different two stable small social networks (A small stable network is consisted on four agents.). Secondly, the authors analyze the network (eight agents) shape which is added one agent to a stable network (seven agents). Thirdly, the authors analyze interpersonal comparison of utility. The “star-network "was not found on the result of interaction among stable two small networks. On the other hand, “decentralized network" was formed from several combination. In case of added one agent to a stable network (seven agents), if the value of “c"(maintenance cost of per a link) was larger, the number of patterns of stable network was also larger. In this case, the authors identified the characteristics of a large stable network. The authors discovered the cases of decreasing personal utility under condition increasing total utility.

Keywords: Social Network, Symmetric Situation, Network Stability, Agent-Based Modeling.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1533

Authors: Nashrawan N. Taha, Andrew M. Cox

Abstract:

In a multi-cultural learning context, where ties are weak and dynamic, combining qualitative with quantitative research methods may be more effective. Such a combination may also allow us to answer different types of question, such as about people’s perception of the network. In this study the use of observation, interviews and photos were explored as ways of enhancing data from social network questionnaires. Integrating all of these methods was found to enhance the quality of data collected and its accuracy, also providing a richer story of the network dynamics and the factors that shaped these changes over time.

Keywords: Mixed Methods, Social Network Analysis, multi-cultural learning, Social Network Dynamics.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1805

Authors: Motonobu Hattori, Rina Suzuki

Abstract:

In neural networks, when new patters are learned by a network, they radically interfere with previously stored patterns. This drawback is called catastrophic forgetting. We have already proposed a biologically inspired dual-network memory model which can much reduce this forgetting for static patterns. In this model, information is first stored in the hippocampal network, and thereafter, it is transferred to the neocortical network using pseudopatterns. Because temporal sequence learning is more important than static pattern learning in the real world, in this study, we improve our conventional  dual-network memory model so that it can deal with temporal sequences without catastrophic forgetting. The computer simulation results show the effectiveness of the proposed dual-network memory model.  

Keywords: Catastrophic forgetting, dual-network, temporal sequences.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1424

Authors: Gunasekaran Raja, Ramkumar Jayaraman, Rajakumar Arul, Kottilingam Kottursamy

Abstract:

Broadband Wireless Network (BWN) is the promising technology nowadays due to the increased number of smartphones. Buffering scheme using network coding considers the reliability and proper degree distribution in Worldwide interoperability for Microwave Access (WiMAX) multi-hop network. Using network coding, a secure way of transmission is performed which helps in improving throughput and reduces the packet loss in the multicast network. At the outset, improved network coding is proposed in multicast wireless mesh network. Considering the problem of performance overhead, degree distribution makes a decision while performing buffer in the encoding / decoding process. Consequently, BuS (Buffer Scheme) based on network coding is proposed in the multi-hop network. Here the encoding process introduces buffer for temporary storage to transmit packets with proper degree distribution. The simulation results depend on the number of packets received in the encoding/decoding with proper degree distribution using buffering scheme.

Keywords: Encoding and decoding, buffer, network coding, degree distribution, broadband wireless networks, multicast.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1740

Authors: Ekachai Phaisangittisagul, Rapeepol Chongprachawat

Abstract:

Obtaining labeled data in supervised learning is often difficult and expensive, and thus the trained learning algorithm tends to be overfitting due to small number of training data. As a result, some researchers have focused on using unlabeled data which may not necessary to follow the same generative distribution as the labeled data to construct a high-level feature for improving performance on supervised learning tasks. In this paper, we investigate the impact of the relationship between unlabeled and labeled data for classification performance. Specifically, we will apply difference unlabeled data which have different degrees of relation to the labeled data for handwritten digit classification task based on MNIST dataset. Our experimental results show that the higher the degree of relation between unlabeled and labeled data, the better the classification performance. Although the unlabeled data that is completely from different generative distribution to the labeled data provides the lowest classification performance, we still achieve high classification performance. This leads to expanding the applicability of the supervised learning algorithms using unsupervised learning.

Keywords: Autoencoder, high-level feature, MNIST dataset, selftaught learning, supervised learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1832

Authors: Abdelhadi Lotfi, Abdelkader Benyettou

Abstract:

In this work, a training algorithm for probabilistic neural networks (PNN) is presented. The algorithm addresses one of the major drawbacks of PNN, which is the size of the hidden layer in the network. By using a cross-validation training algorithm, the number of hidden neurons is shrunk to a smaller number consisting of the most representative samples of the training set. This is done without affecting the overall architecture of the network. Performance of the network is compared against performance of standard PNN for different databases from the UCI database repository. Results show an important gain in network size and performance.

Keywords: Classification, probabilistic neural networks, network optimization, pattern recognition.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1223

Authors: Penghui Zhang, Hua Zhang, Jun-Bo Wang, Cheng Zeng, Zijian Cao

Abstract:

With the development of software-defined networks and programmable data planes, in-band network telemetry (INT) has become an emerging technology in communications because it can get accurate and real-time network information. However, due to the expansion of the network scale, existing telemetry systems, to the best of the authors’ knowledge, have difficulty in meeting the common requirements of low overhead, low latency and full coverage for traffic measurement. This paper proposes a network-wide telemetry system with a low-latency low-overhead path planning (INT-LLPP). This paper builds a mathematical model to analyze the telemetry overhead and latency of INT systems. Then, we adopt a greedy-based path planning algorithm to reduce the overhead and latency of the network telemetry with the full network coverage. The simulation results show that network-wide telemetry is achieved and the telemetry overhead can be reduced significantly compared with existing INT systems. INT-LLPP can control the system latency to get real-time network information.

Keywords: Network telemetry, network monitoring, path planning, low latency.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 253

Authors: Do-Woo Kim, Geon Woo Kim, Jun-Ho Lee, Jong-Wook Han

Abstract:

The home in these days has not one computer connected to the Internet but rather a network of many devices within the home, and that network might be connected to the Internet. In such an environment, the potential for attacks is greatly increased. The general security technology can not apply because of the use of various wired and wireless network, middleware and protocol in digital home environment and a restricted system resource of home information appliances. To offer secure home services home network environments have need of access control for various home devices and information when users want to access. Therefore home network access control for user authorization is a very important issue. In this paper we propose access control model using RBAC in home network environments to provide home users with secure home services.

Keywords: Home network, access control, RBAC, security.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1911

Authors: B. S. Yew, B. L. Ong, R. B. Ahmad

Abstract:

A study on the performance of TCP Vegas versus different TCP variants in homogeneous and heterogeneous wired networks are performed via simulation experiment using network simulator (ns-2). This performance evaluation prepared a comparison medium for the performance evaluation of enhanced-TCP Vegas in wired network and for wireless network. In homogeneous network, the performance of TCP Tahoe, TCP Reno, TCP NewReno, TCP Vegas and TCP SACK are analyzed. In heterogeneous network, the performances of TCP Vegas against TCP variants are analyzed. TCP Vegas outperforms other TCP variants in homogeneous wired network. However, TCP Vegas achieves unfair throughput in heterogeneous wired network.

Keywords: TCP Vegas, Homogeneous, Heterogeneous, WiredNetwork.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1717

Authors: Seongchan Kang, Hongsik Yun, Hyukgil Kim, Minwoo Park

Abstract:

For this study, this researcher conducted a precision network adjustment with QOCA, the precision network adjustment software developed by Jet Propulsion Laboratory, to perform an integrated network adjustment on the Unified Control Points managed by the National Geographic Information Institute. Towards this end, 275 Unified Control Points observed in 2008 were selected before a network adjustment is performed on those 275 Unified Control Points. The RMSE on the discrepancies of coordinates as compared to the results of GLOBK was ±6.07mm along the N axis, ±2.68mm along the E axis and ±6.49mm along the U axis.

Keywords: Network adjustment, QOCA, unified control point.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1848

Authors: Huan Xiong, Ming Chen

Abstract:

In order to research Internet quantificationally and better model the performance of network, this paper proposes a novel AS level network performance model (MNPM), it takes autonomous system (AS) as basic modeling unit, measures E2E performance between any two outdegrees of an AS and organizes measurement results into matrix form which called performance matrix (PM). Inter-AS performance calculation is defined according to performance information stored in PM. Simulation has been implemented to verify the correctness of MNPM and a practical application of MNPM (network congestion detection) is given.

Keywords: AS, network performance, model, metric, congestion.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1408

Authors: M. Mahdavi, E. Mahdavi

Abstract:

In this research, STNEP is being studied considering network adequacy and limitation of investment cost by decimal codification genetic algorithm (DCGA). The goal is obtaining the maximum of network adequacy with lowest expansion cost for a specific investment. Finally, the proposed idea is applied to the Garvers 6-bus network. The results show that considering the network adequacy for solution of STNEP problem is caused that among of expansion plans for a determined investment, configuration which has relatively lower expansion cost and higher adequacy is proposed by GA based method. Finally, with respect to the curve of adequacy versus expansion cost it can be said that more optimal configurations for expansion of network are obtained with lower investment costs.

Keywords: TNEP, Network Adequacy, Investment Cost, GA

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1515

Authors: B. Q. Huang, Tarik Rashid, M-T. Kechadi

Abstract:

this paper presents a multi-context recurrent network for time series analysis. While simple recurrent network (SRN) are very popular among recurrent neural networks, they still have some shortcomings in terms of learning speed and accuracy that need to be addressed. To solve these problems, we proposed a multi-context recurrent network (MCRN) with three different learning algorithms. The performance of this network is evaluated on some real-world application such as handwriting recognition and energy load forecasting. We study the performance of this network and we compared it to a very well established SRN. The experimental results showed that MCRN is very efficient and very well suited to time series analysis and its applications.

Keywords: Gradient descent method, recurrent neural network, learning algorithms, time series, BP

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3043

Authors: Jun-Kwon Jung, Sung-Min Jung, Tae-Kyung Kim, Tai-Myoung Chung

Abstract:

CloudSim is a useful tool to simulate the cloud environment. It shows the service availability, the power consumption, and the network traffic of services on the cloud environment. Moreover, it supports to calculate a network communication delay through a network topology data easily. CloudSim allows inputting a file of topology data, but it does not provide any generating process. Thus, it needs the file of topology data generated from some other tools. The BRITE is typical network topology generator. Also, it supports various type of topology generating algorithms. If CloudSim can include the BRITE, network simulation for clouds is easier than existing version. This paper shows the potential of connection between BRITE and CloudSim. Also, it proposes the direction to link between them.

Keywords: Cloud, simulation, topology, BRITE, network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3778

Authors: Sinarwati Mohamad Suhaili, Naomie Salim, Mohamad Nazim Jambli

Abstract:

Sequence-to-sequence (seq2seq) models augmented with attention mechanisms are increasingly important in automated customer service. These models, adept at recognizing complex relationships between input and output sequences, are essential for optimizing chatbot responses. Central to these mechanisms are neural attention weights that determine the model’s focus during sequence generation. Despite their widespread use, there remains a gap in the comparative analysis of different attention weighting functions within seq2seq models, particularly in the context of chatbots utilizing the Customer Support Twitter (CST) dataset. This study addresses this gap by evaluating four distinct attention-scoring functions—dot, multiplicative/general, additive, and an extended multiplicative function with a tanh activation parameter — in neural generative seq2seq models. Using the CST dataset, these models were trained and evaluated over 10 epochs with the AdamW optimizer. Evaluation criteria included validation loss and BLEU scores implemented under both greedy and beam search strategies with a beam size of k = 3. Results indicate that the model with the tanh-augmented multiplicative function significantly outperforms its counterparts, achieving the lowest validation loss (1.136484) and the highest BLEU scores (0.438926 under greedy search, 0.443000 under beam search, k = 3). These findings emphasize the crucial influence of selecting an appropriate attention-scoring function to enhance the performance of seq2seq models for chatbots, particularly highlighting the model integrating tanh activation as a promising approach to improving chatbot quality in customer support contexts.

Keywords: Attention weight, chatbot, encoder-decoder, neural generative attention, score function, sequence-to-sequence.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 89

Authors: Won Sup Kim, Xue-Mei Cui, Seung Kee Han

Abstract:

We present on the method of inverse coherence matrix for the estimation of network connectivity from multivariate time series of a complex system. In a model system of coupled chaotic oscillators, it is shown that the inverse coherence matrix defined as the inverse of cross coherence matrix is proportional to the network connectivity. Therefore the inverse coherence matrix could be used for the distinction between the directly connected links from indirectly connected links in a complex network. We compare the result of network estimation using the method of the inverse coherence matrix with the results obtained from the coherence matrix and the partial coherence matrix.

Keywords: Chaotic oscillator, complex network, inverse coherence matrix, network estimation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2003

Authors: W.S. Lim, M.V.C. Rao

Abstract:

In this paper, the processing of sonar signals has been carried out using Minimal Resource Allocation Network (MRAN) and a Probabilistic Neural Network (PNN) in differentiation of commonly encountered features in indoor environments. The stability-plasticity behaviors of both networks have been investigated. The experimental result shows that MRAN possesses lower network complexity but experiences higher plasticity than PNN. An enhanced version called parallel MRAN (pMRAN) is proposed to solve this problem and is proven to be stable in prediction and also outperformed the original MRAN.

Keywords: Ultrasonic sensing, target classification, minimalresource allocation network (MRAN), probabilistic neural network(PNN), stability-plasticity dilemma.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1596

Authors: Aboagela Dogman, Reza Saatchi, Samir Al-Khayatt

Abstract:

In this study, a network quality of service (QoS) evaluation system was proposed. The system used a combination of fuzzy C-means (FCM) and regression model to analyse and assess the QoS in a simulated network. Network QoS parameters of multimedia applications were intelligently analysed by FCM clustering algorithm. The QoS parameters for each FCM cluster centre were then inputted to a regression model in order to quantify the overall QoS. The proposed QoS evaluation system provided valuable information about the network-s QoS patterns and based on this information, the overall network-s QoS was effectively quantified.

Keywords: Fuzzy C-means; regression model, network quality of service

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1720