The paper analyzed the relationship between the person’s fourteen characteristic factors and place to visit. The personal factors consist of personality, marital Status, final education, majors, religion, monthly income, commuting means and time, frequency of travel, userage of social media, time spent on social media per day, cultural type. In addition, the analysis was done on which factors have the greatest impact. The analysis involved thirty-four participants and the boosting technique was used as a method of analysis.
{"title":"Analysis of the Correlation Between Personal Factors and Visiting Locations With Boosting Technique","authors":"H. Song, Jiseon Yun","doi":"10.15439/2019F319","DOIUrl":"https://doi.org/10.15439/2019F319","url":null,"abstract":"The paper analyzed the relationship between the person’s fourteen characteristic factors and place to visit. The personal factors consist of personality, marital Status, final education, majors, religion, monthly income, commuting means and time, frequency of travel, userage of social media, time spent on social media per day, cultural type. In addition, the analysis was done on which factors have the greatest impact. The analysis involved thirty-four participants and the boosting technique was used as a method of analysis.","PeriodicalId":168208,"journal":{"name":"2019 Federated Conference on Computer Science and Information Systems (FedCSIS)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129347877","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}
Although the unimodal biometric recognition (such as face and palmprint) has higher convenience, its security is also relatively weak. The recognition accuracy is easy affected by many factors such as ambient light and recognition distance etc. To address this issue, we present a weighted multimodal biometric recognition algorithm with face and palmprint based on histogram of contourlet oriented gradient (HCOG) feature description. We employ the nonsubsampled contour transform (NSCT) to decompose the face and palmprint images, and the HOG method is adopted to extract the feature, which is named as HCOG feature. Then the dimension reduction process is applied on the HCOG feature and a novel weight value computation method is proposed to accomplish the multimodal biometric fusion recognition. Extensive experiments illustrate that our proposed weighted fusion recognition can achieve excellent recognition accuracy rates and outmatches the unimodal biometric recognition methods.
{"title":"Weighted Multimodal Biometric Recognition Algorithm Based on Histogram of Contourlet Oriented Gradient Feature Description","authors":"Xinman Zhang, Dongxu Cheng, Xuebin Xu","doi":"10.15439/2019F178","DOIUrl":"https://doi.org/10.15439/2019F178","url":null,"abstract":"Although the unimodal biometric recognition (such as face and palmprint) has higher convenience, its security is also relatively weak. The recognition accuracy is easy affected by many factors such as ambient light and recognition distance etc. To address this issue, we present a weighted multimodal biometric recognition algorithm with face and palmprint based on histogram of contourlet oriented gradient (HCOG) feature description. We employ the nonsubsampled contour transform (NSCT) to decompose the face and palmprint images, and the HOG method is adopted to extract the feature, which is named as HCOG feature. Then the dimension reduction process is applied on the HCOG feature and a novel weight value computation method is proposed to accomplish the multimodal biometric fusion recognition. Extensive experiments illustrate that our proposed weighted fusion recognition can achieve excellent recognition accuracy rates and outmatches the unimodal biometric recognition methods.","PeriodicalId":168208,"journal":{"name":"2019 Federated Conference on Computer Science and Information Systems (FedCSIS)","volume":"54 8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124696070","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}
The p-median problem is classified as a NP-hard problem, which demands a long time for solution. To increase the use of the method in public management, commercial, military and industrial applications, several heuristic methods has been proposed in literature. In this work, we propose a customized Genetic Algorithm for solving the p-median problem, and we present its evaluation using benchmark problems of OR-library. The customized method combines parameters used in previous studies and introduces the evolution of solutions in stationary mode for solving PMP problems. The proposed Genetic Algorithm found the optimum solution in 37 of 40 instances of p-median problem. The mean deviation from the optimal solution was 0.002% and the mean processing time using CPU core i7 was 17.7s.
{"title":"Customized Genetic Algorithm for Facility Allocation using p-median","authors":"Sergio D. de S. Silva, M. Costa, C. Filho","doi":"10.15439/2019F158","DOIUrl":"https://doi.org/10.15439/2019F158","url":null,"abstract":"The p-median problem is classified as a NP-hard problem, which demands a long time for solution. To increase the use of the method in public management, commercial, military and industrial applications, several heuristic methods has been proposed in literature. In this work, we propose a customized Genetic Algorithm for solving the p-median problem, and we present its evaluation using benchmark problems of OR-library. The customized method combines parameters used in previous studies and introduces the evolution of solutions in stationary mode for solving PMP problems. The proposed Genetic Algorithm found the optimum solution in 37 of 40 instances of p-median problem. The mean deviation from the optimal solution was 0.002% and the mean processing time using CPU core i7 was 17.7s.","PeriodicalId":168208,"journal":{"name":"2019 Federated Conference on Computer Science and Information Systems (FedCSIS)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129101059","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}
Identity verification using biometric methods has been used for many years. A special case is a handwritten signature made on a digital device or piece of paper. For the digital analysis and verification of its authenticity, special methods are needed. Unfortunately, this is a rather complicated task that quite often requires complex processing techmques. In this paper, we propose a system of signatures verification consisting of two stages. In the first one, a signature pattern is created. Thanks to this, the first attempt to verify identity takes place. In the case of approval, the second stage is followed by the processing of a graphic sample contaimng a signature by the convolutional neural network. The proposed techmque has been described, tested and discussed due to its practical use.
{"title":"Signature analysis system using a convolutional neural network","authors":"Alicja Winnicka, K. Kesik, Dawid Połap","doi":"10.15439/2019F28","DOIUrl":"https://doi.org/10.15439/2019F28","url":null,"abstract":"Identity verification using biometric methods has been used for many years. A special case is a handwritten signature made on a digital device or piece of paper. For the digital analysis and verification of its authenticity, special methods are needed. Unfortunately, this is a rather complicated task that quite often requires complex processing techmques. In this paper, we propose a system of signatures verification consisting of two stages. In the first one, a signature pattern is created. Thanks to this, the first attempt to verify identity takes place. In the case of approval, the second stage is followed by the processing of a graphic sample contaimng a signature by the convolutional neural network. The proposed techmque has been described, tested and discussed due to its practical use.","PeriodicalId":168208,"journal":{"name":"2019 Federated Conference on Computer Science and Information Systems (FedCSIS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126411022","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}
R. Leskovský, Erik Kučera, Oto Haffner, Jakub Matisák, D. Rosinová, Erich Stark
The paper demonstrates an application developed to help to evaluate ergonomics of a workplace. Ergonomics of a workplace has enormous impact on employees and their long-term work effectiveness, which causes an interest in this field from employers’ point of view. The paper describes and compares several attitudes companies use to set up and evaluate workplace metrics, potential of virtual reality (VR) in the process, VR application proposal, implementation within Unity 3D engine and results achieved with implementation of this proposed solution. Current approaches also include motion tracking for ergonomics evaluation. These technologies are often far over smaller companies’ budget. Described solution is reasonably priced also for small companies, using cheaper motion capture equipment.
{"title":"A Contribution to Workplace Ergonomics Evaluation Using Multimedia Tools and Virtual Reality","authors":"R. Leskovský, Erik Kučera, Oto Haffner, Jakub Matisák, D. Rosinová, Erich Stark","doi":"10.15439/2019F292","DOIUrl":"https://doi.org/10.15439/2019F292","url":null,"abstract":"The paper demonstrates an application developed to help to evaluate ergonomics of a workplace. Ergonomics of a workplace has enormous impact on employees and their long-term work effectiveness, which causes an interest in this field from employers’ point of view. The paper describes and compares several attitudes companies use to set up and evaluate workplace metrics, potential of virtual reality (VR) in the process, VR application proposal, implementation within Unity 3D engine and results achieved with implementation of this proposed solution. Current approaches also include motion tracking for ergonomics evaluation. These technologies are often far over smaller companies’ budget. Described solution is reasonably priced also for small companies, using cheaper motion capture equipment.","PeriodicalId":168208,"journal":{"name":"2019 Federated Conference on Computer Science and Information Systems (FedCSIS)","volume":"193 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115771119","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}
Miguel Morales Trujillo, Gabriel Alberto García-Mireles
Skills demanded by the IT industry from graduates should be aligned with the curricula of Computer Science undergraduate programs. It is well-known that theoretical knowledge undergraduate students acquire during their studies needs to be complemented with practical experience; therefore, participating in university supported real life projects is a viable option for the students to get prepared for the industry. This paper reports findings from a survey applied to students who had been involved in an industry-based program meant to fulfill their graduation requirements, including the opportunity to develop a capstone project. We gathered their perceptions regarding what they learned during their studies, what they acquired in the industry-based program and what they consider useful for their current jobs. The results show that most topics are aligned between the Bachelor’s degree program and the industry needs, but there is a strong separation in the cognitive levels students achieve at each stage. The paper provides insight into the needs of Computer Science students and contributes to finding ways of increasing undergraduate student satisfaction with skills acquired at university and their application in real contexts.
{"title":"Participating in an Industry Based Social Service Program: a Report of Student Perception of What They Learn and What They Need","authors":"Miguel Morales Trujillo, Gabriel Alberto García-Mireles","doi":"10.15439/2019F279","DOIUrl":"https://doi.org/10.15439/2019F279","url":null,"abstract":"Skills demanded by the IT industry from graduates should be aligned with the curricula of Computer Science undergraduate programs. It is well-known that theoretical knowledge undergraduate students acquire during their studies needs to be complemented with practical experience; therefore, participating in university supported real life projects is a viable option for the students to get prepared for the industry. This paper reports findings from a survey applied to students who had been involved in an industry-based program meant to fulfill their graduation requirements, including the opportunity to develop a capstone project. We gathered their perceptions regarding what they learned during their studies, what they acquired in the industry-based program and what they consider useful for their current jobs. The results show that most topics are aligned between the Bachelor’s degree program and the industry needs, but there is a strong separation in the cognitive levels students achieve at each stage. The paper provides insight into the needs of Computer Science students and contributes to finding ways of increasing undergraduate student satisfaction with skills acquired at university and their application in real contexts.","PeriodicalId":168208,"journal":{"name":"2019 Federated Conference on Computer Science and Information Systems (FedCSIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130947942","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}
Environmental sound classification has received more attention in recent years. Analysis of environmental sounds is difficult because of its unstructured nature. However, the presence of strong spectro-temporal patterns makes the classification possible. Since LSTM neural networks are efficient at learning temporal dependencies we propose and examine a LSTM model for urban sound classification. The model is trained on magnitude mel-spectrograms extracted from UrbanSound8K dataset audio. The proposed network is evaluated using 5-fold cross-validation and compared with the baseline CNN. It is shown that the LSTM model outperforms a set of existing solutions and is more accurate and confident than the CNN.
{"title":"Urban Sound Classification using Long Short-Term Memory Neural Network","authors":"Yurij Lezhenin, N. Bogach, Evgeny Pyshkin","doi":"10.15439/2019F185","DOIUrl":"https://doi.org/10.15439/2019F185","url":null,"abstract":"Environmental sound classification has received more attention in recent years. Analysis of environmental sounds is difficult because of its unstructured nature. However, the presence of strong spectro-temporal patterns makes the classification possible. Since LSTM neural networks are efficient at learning temporal dependencies we propose and examine a LSTM model for urban sound classification. The model is trained on magnitude mel-spectrograms extracted from UrbanSound8K dataset audio. The proposed network is evaluated using 5-fold cross-validation and compared with the baseline CNN. It is shown that the LSTM model outperforms a set of existing solutions and is more accurate and confident than the CNN.","PeriodicalId":168208,"journal":{"name":"2019 Federated Conference on Computer Science and Information Systems (FedCSIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131296649","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}
In the paper, a new multi-level hybrid method of community detection combining a density-based clustering with a label propagation method is proposed. Many algorithms have been applied to preprocess, visualize, cluster, and interpret the data describing customer behavior, among others DBSCAN, RFM, k-NN, UMAP, LPA. In the paper, two key algorithms have been detailed: DBSCAN and LPA. DBSCAN is a density-based clustering algorithm. However, managers usually find the clustering results too difficult to interpret and apply. To enhance the business value of clustering and create customer communities, the label propagation algorithm (LPA) has been proposed due to its quality and low computational complexity. The approach is validated on real life marketing database using advanced analytics platform Upsaily.
{"title":"An Approach to Customer Community Discovery","authors":"J. Korczak, Maciej Pondel, Wiktor Sroka","doi":"10.15439/2019F308","DOIUrl":"https://doi.org/10.15439/2019F308","url":null,"abstract":"In the paper, a new multi-level hybrid method of community detection combining a density-based clustering with a label propagation method is proposed. Many algorithms have been applied to preprocess, visualize, cluster, and interpret the data describing customer behavior, among others DBSCAN, RFM, k-NN, UMAP, LPA. In the paper, two key algorithms have been detailed: DBSCAN and LPA. DBSCAN is a density-based clustering algorithm. However, managers usually find the clustering results too difficult to interpret and apply. To enhance the business value of clustering and create customer communities, the label propagation algorithm (LPA) has been proposed due to its quality and low computational complexity. The approach is validated on real life marketing database using advanced analytics platform Upsaily.","PeriodicalId":168208,"journal":{"name":"2019 Federated Conference on Computer Science and Information Systems (FedCSIS)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127514548","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}
Artur Karczmarczyk, Jarosław Jankowski, J. Wątróbski
Spreading of information within social media and techniques related to viral marketing take more and more attention from companies focused on targeting audiences within electronic systems. Recent years resulted in extensive research centered around spreading models, selection of initial nodes within networks and identification of campaign characteristics affecting the assumed goals. While social networks are usually based on complex structures and high number of users, the ability to perform detailed analysis of mechanics behind the spreading processes is very limited. The presented study shows an approach for selection of campaign parameters with the use of network samples and theoretical models. Instead of processing simulations on large network, smaller samples and theoretical networks are used. Results showed that knowledge derived from relatively smaller structures is helpful for initialization of spreading processes within the target network of larger size. Apart from agent based modeling, multi-criteria methods were used for evaluation of results from the perspective of costs and performance.
{"title":"Multi-criteria approach to viral marketing campaign planning in social networks, based on real networks, network samples and synthetic networks","authors":"Artur Karczmarczyk, Jarosław Jankowski, J. Wątróbski","doi":"10.15439/2019F199","DOIUrl":"https://doi.org/10.15439/2019F199","url":null,"abstract":"Spreading of information within social media and techniques related to viral marketing take more and more attention from companies focused on targeting audiences within electronic systems. Recent years resulted in extensive research centered around spreading models, selection of initial nodes within networks and identification of campaign characteristics affecting the assumed goals. While social networks are usually based on complex structures and high number of users, the ability to perform detailed analysis of mechanics behind the spreading processes is very limited. The presented study shows an approach for selection of campaign parameters with the use of network samples and theoretical models. Instead of processing simulations on large network, smaller samples and theoretical networks are used. Results showed that knowledge derived from relatively smaller structures is helpful for initialization of spreading processes within the target network of larger size. Apart from agent based modeling, multi-criteria methods were used for evaluation of results from the perspective of costs and performance.","PeriodicalId":168208,"journal":{"name":"2019 Federated Conference on Computer Science and Information Systems (FedCSIS)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130696471","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}
Day for day it becomes easier to temper digital images. Thus, people are in need of various forgery image detection. In this paper, we present forgery image detection techniques for two of the most common image tampering techniques; copy-move and splicing. We use match points technique after feature extraction process using SIFT and SURF. For splicing detection, we extracted the edges of the integral images of $Y, C_{b}$, and $C_{r}$ image components. GLCM is applied for each edge integral image and the feature vector is formed. The feature vector is then fed to a SVM classifier. For the copy-move, the results show that SURF feature extraction can be more efficient than SIFT, where we achieved 80% accuracy of detecting tempered images. On the other hand, processing the image in $YC _{b}C_{r}$ color model is found to give promising results in splicing image detection. We have achieved 99% true positive rate for detecting splicing images.
{"title":"Robust Image Forgery Detection Using Point Feature Analysis","authors":"Youssef William, S. Safwat, M. A. Salem","doi":"10.15439/2019F227","DOIUrl":"https://doi.org/10.15439/2019F227","url":null,"abstract":"Day for day it becomes easier to temper digital images. Thus, people are in need of various forgery image detection. In this paper, we present forgery image detection techniques for two of the most common image tampering techniques; copy-move and splicing. We use match points technique after feature extraction process using SIFT and SURF. For splicing detection, we extracted the edges of the integral images of $Y, C_{b}$, and $C_{r}$ image components. GLCM is applied for each edge integral image and the feature vector is formed. The feature vector is then fed to a SVM classifier. For the copy-move, the results show that SURF feature extraction can be more efficient than SIFT, where we achieved 80% accuracy of detecting tempered images. On the other hand, processing the image in $YC _{b}C_{r}$ color model is found to give promising results in splicing image detection. We have achieved 99% true positive rate for detecting splicing images.","PeriodicalId":168208,"journal":{"name":"2019 Federated Conference on Computer Science and Information Systems (FedCSIS)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115912248","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：