Hoda Memarzadeh, Nasser Ghadiri, Sara Parikhah Zarmehr
{"title":"A Graph Database Approach for Temporal Modeling of Disease Progression","authors":"Hoda Memarzadeh, Nasser Ghadiri, Sara Parikhah Zarmehr","doi":"10.1109/ICCKE.2018.8566311","DOIUrl":"https://doi.org/10.1109/ICCKE.2018.8566311","url":null,"abstract":"The high cost of managing chronic diseases for individuals and governments, as well as the negative impact on the quality of life, highlights the importance of controlling and preventing the chronic disease progression. Understanding the disease progression model is one of the first steps, which can lead to more effective planning for interventions. Most of the different approaches for statistical modeling of disease progression work with the graph. On the other hand longitudinal medical data could be represented in the form of a graph and modeling them in this way has a great deal of potential for analyzing and tracking medical event. Data structures, data model features, query facilities and special commands in graph database for traversing and detection patterns could be useful for building summarized information based on transitions between different stages of a particular disease in individual graphs. Given the fact that clinical data is collected at different times, software and formats, there is a need for a flexible framework for data linkage. Use of graph databases brings this flexibility into account and provide a highly scalable framework for data integrating and linkage. In this study, at first simple medical observations related to patients with varying degrees of Alzheimer's disease stored in a graph database (Neo4j) and then by reviewing the capabilities of this environment in building transition graph of different stages of the disease, suggestions for the model development with more details were presented.","PeriodicalId":283700,"journal":{"name":"2018 8th International Conference on Computer and Knowledge Engineering (ICCKE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116516348","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":"Robust Two Stage Unsupervised Metric Learning for Domain Adaptation","authors":"Samaneh Azarbarzin, F. Afsari","doi":"10.1109/ICCKE.2018.8566472","DOIUrl":"https://doi.org/10.1109/ICCKE.2018.8566472","url":null,"abstract":"Most commonly used metric learning procedures suppose that the input feature space and domain of the training and test data are identical. In such cases these algorithms cannot improve target learning problems. This paper presents a robust distance metric for domain adaptation in two stages. At first stage both source and target features are transferred to a newly found latent feature space, which minimizes the difference between domains as well as the data properties are preserved. Then in the second stage, the desired metric is learned with a marginalized denoising strategy and the low-rank constraint. To show the superiority and power of the proposed method it is tested on distinct kinds of cross-domain image categorization datasets and the results prove that our approach remarkably exceeds other existing domain adaptation algorithms in the classification tasks.","PeriodicalId":283700,"journal":{"name":"2018 8th International Conference on Computer and Knowledge Engineering (ICCKE)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129640026","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":"The Study on Quranic Surahs' Topic Sameness Using NLP Techniques","authors":"Ehsan Khadangi, M. M. Fazeli, Amin Shahmohammadi","doi":"10.1109/ICCKE.2018.8566248","DOIUrl":"https://doi.org/10.1109/ICCKE.2018.8566248","url":null,"abstract":"Study of the structured-ness of Quranic surahs has attracted the attention of some researchers in recent years. One of the theories herein is the theory of Topic Sameness which acknowledges that the inner elements of surahs have tight relationship with each other and that each surah of Quran has formed on a single core topic. In this paper, we intend to study the topic sameness in Quranic surahs using natural language processing methods. In this regard, based on the two methods of word2vec and Roots' accompaniment in Verses, the similarity of Quranic roots is calculated. Then, the amount of similarity between surahs' title and the concepts within the surahs is studied. Afterwards, the amount of similarity of the concepts within chapters to each other is calculated and compared with the random mode. The results show that the choice of the surah's title is based on rational logic, and that could not have been done by the ordinary public of the early Islamic era. In addition, the surahs hold the inner coherence between the concepts so that they have formed on a single topic or a few topics tightly related to each other.","PeriodicalId":283700,"journal":{"name":"2018 8th International Conference on Computer and Knowledge Engineering (ICCKE)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126813002","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 Combination of Bees and Firefly Algorithm to Optimize Continuous Problems","authors":"Jafar Gholami, Somayeh Mohammadi","doi":"10.1109/ICCKE.2018.8566263","DOIUrl":"https://doi.org/10.1109/ICCKE.2018.8566263","url":null,"abstract":"Although Bees algorithm (BA) has presented great performance so far in optimizing different problems, it still has weak convergence rate in some optimizations. Therefore, in this investigation in order to resolve its weakness and boost the exploitation of algorithm, it was hybridized by Firefly algorithm (FA). The hybrid bees and firefly algorithm (BA-FA) was implemented on twelve numerical benchmark functions to evaluate it. The BA-FA was also compared with four famous algorithms, including particle swarm optimization (PSO), invasive weed optimization (IWO) and so on. The results showed that BA-FA had the best performance and convergence in optimizing the mentioned functions.","PeriodicalId":283700,"journal":{"name":"2018 8th International Conference on Computer and Knowledge Engineering (ICCKE)","volume":"193 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131183523","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":"Toward a Progressive Optimization: Application to Control Design","authors":"Jalaeian-F. Mohsen, S. M. Ahmadi","doi":"10.1109/ICCKE.2018.8566621","DOIUrl":"https://doi.org/10.1109/ICCKE.2018.8566621","url":null,"abstract":"This paper outlines a new approach to optimal control using a firefly progressive algorithm. Having the proposed progressive control method allowed us to solve several problems of offline and online optimization algorithms. To make it clear, offline optimization algorithms suffer from a time-consuming procedure before control runtime, and inactivity during the control runtime. At the other end of the extreme, online optimization algorithms pay more attention to the exploitation rather than the exploration, and therefore trapping more in the local minima. To overcome these concerns, the proposed scheme suggests applying an offline optimization algorithm, as an online progressive optimizer during the runtime of the control system, as a novel way. Here, the firefly algorithm is utilized as the progressive optimization method. To validate the highperformance of the proposed scheme, a numerical example has been conducted.","PeriodicalId":283700,"journal":{"name":"2018 8th International Conference on Computer and Knowledge Engineering (ICCKE)","volume":"51 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113972417","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}
Fatemeh Azizmalayeri, S.M. Moein Peyghambarzadeh, Hassan Khotanlou, Amir Salarpour
{"title":"Kernel Correlation Based CNN for Point Cloud Classification Task","authors":"Fatemeh Azizmalayeri, S.M. Moein Peyghambarzadeh, Hassan Khotanlou, Amir Salarpour","doi":"10.1109/ICCKE.2018.8566273","DOIUrl":"https://doi.org/10.1109/ICCKE.2018.8566273","url":null,"abstract":"3D data provides rich information compared to 2D images in machine vision applications. One important type of 3D data is point cloud due to its availability and flexibility. With the success of deep learning methods in almost every machine vision task, the focus of researches in point cloud processing has shifted from hand crafted shape descriptors to learned ones. Convolutional neural networks among all deep learning methods are very popular in image analysis fields, but they cannot be used for point cloud because of point cloud's irregular format and unordered instinct. In this paper we adapted kernel correlation, a technique widely used in point clouds registration field, in order to develop a CNN -like method which extracts local information from point clouds. We propose a new way of measuring similarity between a kernel and the input point cloud data, our method demonstrates competitive results for point clouds classification task.","PeriodicalId":283700,"journal":{"name":"2018 8th International Conference on Computer and Knowledge Engineering (ICCKE)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128748518","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 New Semi-Supervised Method for Network Traffic Classification Based on X-Means Clustering and Label Propagation","authors":"Fakhroddin Noorbehbahani, Sadeq Mansoori","doi":"10.1109/ICCKE.2018.8566608","DOIUrl":"https://doi.org/10.1109/ICCKE.2018.8566608","url":null,"abstract":"Network traffic classification is an essential requirement for network management. Various approaches have been developed for network traffic classification. Traditional approaches such as analysis of port number or payload have some limitations. For example, using port numbers for traffic classification fails if an application uses dynamic port number or applies encryption methods. To address such limitations, modern traffic classification methods employ machine learning techniques. However, machine learning-based traffic classification needs a large labeled data to extract accurate classification model which is expensive and time-consuming. To overcome this issue, we propose a new semi-supervised method for traffic classification based on x-means clustering algorithm and a new label propagation technique. The accuracy of the proposed method tested on Moore's dataset is 0.95 that shows its effectiveness for learning a network traffic classifier using a limited labeled data.","PeriodicalId":283700,"journal":{"name":"2018 8th International Conference on Computer and Knowledge Engineering (ICCKE)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133888401","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":"Soft Meta-Cognitive Neural Network for Classification Problems","authors":"Maedeh Kafiyan, M. Rouhani","doi":"10.1109/ICCKE.2018.8566359","DOIUrl":"https://doi.org/10.1109/ICCKE.2018.8566359","url":null,"abstract":"Classification problems in a sequential framework is considered as an important field in pattern recognition. One of the main concerns in these types of problems is overtraining. In metacognitive neural networks (McNN), overtraining could be avoided by using the confidence of classifier (CoC) measure, which assigns a value between [0], [1] to class label. In this paper, a more accurate measure of CoC for McNN is presented. In addition, the hinge loss function which has no particular probabilistic interpretation is replaced by cross-entropy loss, which the output layer of the Soft meta-cognitive neural network (SMcNN) is a SoftMax layer. The classification performance is improved by applying the proposed SMcNN to well-known vCI datasets and comparing the results to McNN, SVM, and some other classifiers.","PeriodicalId":283700,"journal":{"name":"2018 8th International Conference on Computer and Knowledge Engineering (ICCKE)","volume":"134 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115807055","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":"Test Model Generation Using Equivalence Partitioning","authors":"Sorour Jahanbin, B. Zamani","doi":"10.1109/ICCKE.2018.8566335","DOIUrl":"https://doi.org/10.1109/ICCKE.2018.8566335","url":null,"abstract":"Model transformation, in a simple definition, is a program that accepts a model as input and generates another model as output. Model transformations are the cornerstone of model-driven engineering (MDE), hence testing them and ensuring the correctness of their implementation is a critical task. A challenging aspect of testing model transformations is to generate test models that both conform to their meta-model and satisfy the defined constraints. There exist several solutions for generating test models. Epsilon Model Generation (EMG) is a language for generating appropriate test models. EMG uses random operations for producing test models, hence it is possible that some tests have the same structure and the same value, i.e., they are redundant. In this paper, we propose an approach for generating appropriate test models, i.e., test models which are valuable from the tester's point of view. In this approach, the tester specifies the number of model elements that should be generated in the test model, as well as how they are linked. Our approach is based on the idea of enriching the EMG language with equivalence partitioning technique. The idea of partitioning is that testing a member in an equivalence class is as good as testing the whole class. We have evaluated the proposed method via a case study. The results show the superiority of the proposed approach over EMG.","PeriodicalId":283700,"journal":{"name":"2018 8th International Conference on Computer and Knowledge Engineering (ICCKE)","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123638450","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":"Modeling Temporal Dynamics of User Preferences in Movie Recommendation","authors":"Hamidreza Tahmasbi, Mehrdad Jalali, H. Shakeri","doi":"10.1109/ICCKE.2018.8566316","DOIUrl":"https://doi.org/10.1109/ICCKE.2018.8566316","url":null,"abstract":"Users in movie recommender systems are likely to change their preferences over time. Modelling the temporal dynamics of user preferences is essential for improving the recommendation accuracy. In this paper, we propose an approach to model temporal dynamics of user preferences in movie recommendation systems based on a coupled tensor factorization framework. We weigh the past user preferences and decrease their importance gradually by introducing an individual time decay factor for each user according to the rate of his preference dynamics. We exploit users' demographics as well as the extracted similarities among users over time, aiming to enhance the prior knowledge about user preference dynamics, in addition to the past weighted user preferences to generate movie recommendations. Our experiments on the public benchmark dataset, MovieLens show that our model outperforms other competitive methods and is more capable of alleviating the problems of cold-start and data sparsity.","PeriodicalId":283700,"journal":{"name":"2018 8th International Conference on Computer and Knowledge Engineering (ICCKE)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129248716","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}