2021 IEEE 15th International Symposium on Applied Computational Intelligence and Informatics (SACI)最新文献

{"title":"Soft-Clipping Swish: A Novel Activation Function for Deep Learning","authors":"Marina Adriana Mercioni, S. Holban","doi":"10.1109/SACI51354.2021.9465622","DOIUrl":"https://doi.org/10.1109/SACI51354.2021.9465622","url":null,"abstract":"This study aims to contribute to the improvement of the network’s performance through developing a novel activation function. Over time, many activation functions have been proposed in order to solve the issues of the previous functions. We note here more than 50 activation functions that have been proposed, some of them being very popular such as sigmoid, Rectified Linear Unit (ReLU), Swish, Mish but not only. The main idea of this study that stays behind our proposal is a simple one, based on a very popular function called Swish, which is a composition function, having in its componence sigmoid function and ReLU function. Starting from this activation function we decided to ignore the negative region in the way the Rectified Linear Unit does but being different than that one mentioned through a nonlinear curve assured by the Swish positive region. The idea has been come up from a current function called Soft Clipping. We tested this proposal on more datasets in Computer Vision on classification tasks showing its high potential, here we mention MNIST, Fashion-MNIST, CIFAR-10, CIFAR-100 using two popular architectures: LeNet-5 and ResNet20 version 1.","PeriodicalId":321907,"journal":{"name":"2021 IEEE 15th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116785157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparative electrical impedance spectroscopy study of graphene and indium tin oxide electrodes during low frequency measurements","authors":"A. Hencz, Gábor S. Szilágyi, Nina Gyorfi, Kristof Tenzlinger, A. Széchenyi, Á. Odry, P. Odry, Z. Karádi, Zoltán Vízvári, Attila Tóth","doi":"10.1109/SACI51354.2021.9465546","DOIUrl":"https://doi.org/10.1109/SACI51354.2021.9465546","url":null,"abstract":"The four-electrode system is often applied to examine the dielectric properties of biological samples in low frequency range because the four-electrode method can be used to avoid measurement inaccuracies due to electrode polarization. This study is aiming to create four-electrode sensors that can provide reliable data about measuring cells even at low frequencies. Graphene and indium tin oxide (ITO) covered macroelectrodes were fabricated and compared, using electrical impedance spectroscopy (EIS). Impedance measurements were performed with solution series of NaCl. The impedance values of NaCl concentrations were not significantly different for the graphene and ITO electrodes but the standard deviation of data were larger for ITO electrodes than for graphene electrodes. Based on our investigations, the graphene-coated four-electrode sensor can provide cost-effective and sensitive measurements in the study of alpha and beta dispersions of cells.","PeriodicalId":321907,"journal":{"name":"2021 IEEE 15th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129716068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Novel Hybrid AF/DF Cooperative Communication Scheme for Power Domain NOMA","authors":"M. Ajmal, M. Zeeshan","doi":"10.1109/SACI51354.2021.9465621","DOIUrl":"https://doi.org/10.1109/SACI51354.2021.9465621","url":null,"abstract":"Future wireless communication demand high data rates, low latency, high accuracy, better quality of service and much more. These demands raise challenges in wireless communication system because signal is traveling through medium where signal quality is degraded by random fades, diffraction, noise and other performance reducing phenomenon. To meet the demands, we need technological advancements, so we can alleviate these challenges. NOMA is emerging technique of 5G that ensures better performance than traditional multiple access techniques. In a typical NOMA system, users individually communicate with base station and treat any other user signal as noise. Here the important point is that NOMA strong users have information of all weak users. In this paper, we propose a novel hybrid cooperative communication scheme for multiuser power domain NOMA based on amplify-and-forward (AF) and decode-and-forward (DF) over strong user to improve performance of weak user in cellular system. We exploited the key feature of NOMA that states that strong user has prior knowledge of weak user with bad channel condition. Strong user is used as relay to improve performance of weak user via cooperative communication. BER Curves analysis proved that performance of weak user laying over the edge in cellular system can be improved via cooperative communication.","PeriodicalId":321907,"journal":{"name":"2021 IEEE 15th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129344180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Synthesis of Merging algorithms on binary trees using multisets in Theorema","authors":"I. Dramnesc, T. Jebelean","doi":"10.1109/SACI51354.2021.9465619","DOIUrl":"https://doi.org/10.1109/SACI51354.2021.9465619","url":null,"abstract":"We present the principles and the experiments of deduction based synthesis of merging algorithms for binary trees. Merging is an auxiliary function used in certain algorithms for sorting of binary trees, however we also address concatenation of unsorted trees (the result will also be in general not sorted) as well as merging of sorted trees (both inputs as well as the output are sorted). We follow the classical approach to deductive synthesis: the algorithm is extracted from the proof of a synthesis conjecture, which asserts that the merged object exists. The novelty of our approach consists in using multisets for expressing the fact that two trees have the same content, which also allows us to develop new powerful proof techniques, mostly by using induction based on the Noetherian ordering among trees induced by the strict inclusion of the corresponding multisets. As the synthesis proofs proceed on different alternatives, several versions of the algorithms are produced. The experiments result in 24 concatenation algorithms, 4 merging algorithms, as well as some sub-auxiliary algorithms for insertion of an element into the tree and for splitting a tree into two sub-trees having the elements smaller, respectively bigger, than a certain value.","PeriodicalId":321907,"journal":{"name":"2021 IEEE 15th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"115 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116272930","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Gastrointestinal microbiome depletion modifies behavioral processes without changing body composition","authors":"Kitti Mintál, Zoltán Vízvári, E. Hormay, Anita Kovács, A. Bufa, T. Marosvölgyi, L. Lénárd, Z. Karádi, Attila Tóth","doi":"10.1109/SACI51354.2021.9465575","DOIUrl":"https://doi.org/10.1109/SACI51354.2021.9465575","url":null,"abstract":"In recent times, our knowledge on the gastrointestinal microbiota has considerably improved. Dysfunctions of the gut microbiome have been shown to induce modifications of peripheral and central regulatory processes, ultimately leading to behavioral changes that cause functional deficits in brain functions. Therefore, the first aim of the present study was to generate massive dysbiosis of the gut microbiome and to investigate these effects on behavioral responses in adulthood. To determine the impact of the alterations on the behavior we used adult male Wistar rats. Animals have been divided into two groups – the first one was an antibiotics treated-and the other one was a control group. As antibiotic treatment, rats were given broad spectrum antibiotic mixture for 4 weeks. Substances were dissolved in the animals’ drinking water. Following the intestinal microbiome depletion. Open filed test (OFT) was conducted. Throughout the whole experiment, body weight, food and water consumptions were daily monitored, and faecal samples were collected for later determination of the short chain fatty acid (SCFA) levels. The findings demonstrated significant group-differences such as abnormal behavioral and metabolic phenomena identified among the antibiotic treated animals unraveled in the behavioral test as well as in the short chain fatty acid analyses. Our results reinforce that alterations of the microbiota play important modulating role in the central regulatory processes. However, it is uncertain whether the antibiotic-induced dysbiosis directly triggers the altered behavioral phenomena, or these might have caused the change of body composition which itself could define the behavior. Exploring this question will be our next step to determine how the alterations of microbiome can affect behavioral patterns. We aim to introduce a non-invasive body composition measurement technique to explore this question. This method based on the electrical impedance spectrum analysis. By means of this new four-electrode ultra-low frequency bioimpedance measurement technology, we can carry out high-precision full-body measurements or we could correlate bioimpedance measurement data to various parts of the body if it required. By using the bioimpedance technology, we did not observe significant body composition differences between the two groups. These results confirm our hypothesis that antibiotic-induced dysbiosis directly influences the control of behavioral processes.","PeriodicalId":321907,"journal":{"name":"2021 IEEE 15th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114796397","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An intelligent traffic congestion detection approach based on fuzzy inference system","authors":"Mehran Amini, Miklós F. Hatwágner, G. Mikulai, L. Kóczy","doi":"10.1109/SACI51354.2021.9465637","DOIUrl":"https://doi.org/10.1109/SACI51354.2021.9465637","url":null,"abstract":"Traffic congestion causes significant economic and social consequences. Instant detection of vehicular traffic breakdown has a pivotal role in intelligent transportation engineering. Common traffic estimators and predictors systems need traffic observations to be classified in their binary-set-nature computation methods which are unable to be an effective base for traffic modeling, since they are defined by precise and deterministic characteristics while traffic is known to be a highly complex and nonlinear system, which may be prescribed by uncertain models containing vague properties. This study aims at applying a new fuzzy inference model for predicting the level of congestion in such heterogeneous and convoluted networks, where the paucity of accurate and real-time data can cause problems in interpreting the whole system state by conventional quantitative techniques. The proposed fuzzy inference model is based on real data extracted from Hungarian network of freeways. As input variables traffic flow and approximate capacity of each segment are considered and level of congestion is regarded as output variable. In the model, a total number of 75 rules were developed on the basis of available datasets, percentile distribution, and experts’ judgments. Designed model and analyzing steps are simulated and proven by Matlab fuzzy logic toolbox. The results illustrate correlations and relationships among input variables with predicting the level of congestion based on available resources. Furthermore, performed analyses beside their tractability in dealing with ambiguity and subjectivity are aligned with intelligent traffic modeling purposes in designing traffic breakdown-related alert or early warning systems, infrastructure and services planning, and sustainability development.","PeriodicalId":321907,"journal":{"name":"2021 IEEE 15th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"71 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128011181","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predict Stock Market Prices with Recurrent Neural Networks using NASDAQ Data Stream","authors":"Adam Kovacs, Bence Bogdandy, Zsolt Tóth","doi":"10.1109/SACI51354.2021.9465634","DOIUrl":"https://doi.org/10.1109/SACI51354.2021.9465634","url":null,"abstract":"Prediction and modeling of stock market changes attract not only economists and other scientific professionals, but the general public as well. With the rise of blockchain, and cryptocurrencies, the interest in stock trading has surged. Stock prices are precisely recorded in frequent fixed intervals, and this data is publicly available. Due to the outstanding performance of recurrent neural networks in sequential data modeling, recurrent networks can be applied to model the stock market. Although accurate prediction of the changes is not possible due to the stock market’s highly stochastic stochastic nature, a recurrent neural network could give a good approximation of trends. National Association of Securities Dealers Automated Quotations publishes stock values every few minutes, which data stream was used to model and predict changes. This paper presents a proof of concept implementation of a stock market price prediction system using recurrent neural networks and a continuous data stream.","PeriodicalId":321907,"journal":{"name":"2021 IEEE 15th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"124 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133564779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Simplification on mobile devices reduces personal (cyber) safety","authors":"Anikó Szarvák, Valéria Póser, M. Kozlovszky","doi":"10.1109/SACI51354.2021.9465589","DOIUrl":"https://doi.org/10.1109/SACI51354.2021.9465589","url":null,"abstract":"Due pandemic reasons, the phishing attack numbers are 6 times bigger than normally, approximately. If we assume the defense stays in the same level, this means 6 times more loss than before. Phishing attacks are targeting not just companies but also private individuals, however actual data is not available on individual attack, so the real effect on individuals, their cumulative loss with business is unknown. This study targets the review of mobile attacks and available prevention assets and materials. This paper collects evidences on how simplified – mobile – interfaces according to PC reduces the opportunity to identify cyberattacks as phishing attack. The goal of this document is to point out minor differences between PC-s and mobiles and their user behavior. Mobile devices are widely used and widely targeted by attackers and mobile devices in contrast with PC provides fewer opportunity and feedback to users to avoid attacks.","PeriodicalId":321907,"journal":{"name":"2021 IEEE 15th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122050593","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reliability of a Transport Node based on Fuzzy Fault Tree Analysis","authors":"J. Z. Szabó, P. Bakucz","doi":"10.1109/SACI51354.2021.9465556","DOIUrl":"https://doi.org/10.1109/SACI51354.2021.9465556","url":null,"abstract":"In this work to be introduced a fuzzy fault tree real time algorithm to identify reliability of highly autonomous driving (HAD) redundant system of a traffic node. The fuzzyfication of the fault tree procedure was realized by expressing the basic events with a membership function. Logical gates had to be generalized based on incoming fuzzy events. Node dynamics were discretized in the form of a Laplace graph. The edges of the graph as fuzzy base events and the nodes of the graph as logical gates determine the reliability of the node.","PeriodicalId":321907,"journal":{"name":"2021 IEEE 15th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127292694","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reproducibility analysis of mobile robots movement by IMU sensor measurements","authors":"I. Kecskés, Robert Bacsa, P. Odry, Zoltán Vízvári, Attila Tóth","doi":"10.1109/SACI51354.2021.9465627","DOIUrl":"https://doi.org/10.1109/SACI51354.2021.9465627","url":null,"abstract":"Mobile robot control mechanism required robust solution due to the wide range of movement possibilities and environmental extremums. The reproducibility analysis helps to identify the uncertainty related to movement scenario and measurement unit, such as Inertial Measurement Unit (IMU) sensor. In this paper the reproducibility of accelerometer and gyroscope sensor signals were evaluated on repeated robot movements, which were collected on two mobile robots: a hexapod robot during walking and a quadcopter drone robot during flying. The scenario time scale variability was diminished by dynamic time warping in case of drone robot. The standard deviation of variability from mean signal reach 20-50% compared to the standard deviation of the typical signal.","PeriodicalId":321907,"journal":{"name":"2021 IEEE 15th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127366525","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}