Original/Review Paper
H.3. Artificial Intelligence
Z. Karimi Zandian; M. R. Keyvanpour
Abstract
Fraud detection is one of the ways to cope with damages associated with fraudulent activities that have become common due to the rapid development of the Internet and electronic business. There is a need to propose methods to detect fraud accurately and fast. To achieve to accuracy, fraud detection methods ...
Read More
Fraud detection is one of the ways to cope with damages associated with fraudulent activities that have become common due to the rapid development of the Internet and electronic business. There is a need to propose methods to detect fraud accurately and fast. To achieve to accuracy, fraud detection methods need to consider both kind of features, features based on user level and features based on network level. In this paper a method called MEFUASN is proposed to extract features that is based on social network analysis and then both of obtained features and features based on user level are combined together and used to detect fraud using semi-supervised learning. Evaluation results show using the proposed feature extraction as a pre-processing step in fraud detection improves the accuracy of detection remarkably while it controls runtime in comparison with other methods.
Original/Review Paper
H.3.15.1. Adaptive hypermedia
M. Tahmasebi; F. Fotouhi; M. Esmaeili
Abstract
Personalized recommenders have proved to be of use as a solution to reduce the information overload problem. Especially in Adaptive Hypermedia System, a recommender is the main module that delivers suitable learning objects to learners. Recommenders suffer from the cold-start and the sparsity ...
Read More
Personalized recommenders have proved to be of use as a solution to reduce the information overload problem. Especially in Adaptive Hypermedia System, a recommender is the main module that delivers suitable learning objects to learners. Recommenders suffer from the cold-start and the sparsity problems. Furthermore, obtaining learner’s preferences is cumbersome. Most studies have only focused on similarity between the interest profile of a user and those of others. However, it can lead to the gray-sheep problem, in which users with consistently different opinions from the group do not benefit from this approach. On this basis, matching the learner’s learning style with the web page features and mining specific attributes is more desirable. The primary contribution of this research is to introduce a feature-based recommender system that delivers educational web pages according to the user's individual learning style. We propose an Educational Resource recommender system which interacts with the users based on their learning style and cognitive traits. The learning style determination is based on Felder-Silverman theory. Furthermore, we incorporate all explicit/implicit data features of a page and the elements contained in them that have an influence on the quality of recommendation and help the system make more effective recommendations.
Original/Review Paper
B.3. Communication/Networking and Information Technology
Z. Shaeiri; J. Kazemitabar; Sh. Bijani; M. Talebi
Abstract
As fraudsters understand the time window and act fast, real-time fraud management systems becomes necessary in Telecommunication Industry. In this work, by analyzing traces collected from a nationwide cellular network over a period of a month, an online behavior-based anomaly detection system is provided. ...
Read More
As fraudsters understand the time window and act fast, real-time fraud management systems becomes necessary in Telecommunication Industry. In this work, by analyzing traces collected from a nationwide cellular network over a period of a month, an online behavior-based anomaly detection system is provided. Over time, users' interactions with the network provides a vast amount of usage data. These usage data are modeled to profiles by which users can be identified. A statistical model is proposed that allocate a risk number to each upcoming record which reveals deviation from the normal behavior stored in profiles. Based on the amount of this deviation a decision is made to flag the record as normal or anomaly. If the activity is normal the associated profile is updated; otherwise the record is flagged as anomaly and it will be considered for further investigation. For handling the big data set and implementing the methodology we have used the Apache Spark engine which is an open source, fast and general-purpose cluster computing system for big data handling and analyzes. Experimental results show that the proposed approach can perfectly detect deviations from the normal behavior and can be exploited for detecting anomaly patterns.
Original/Review Paper
H.6.2.2. Fuzzy set
Sh. Asadi; Seyed M. b. Jafari; Z. Shokrollahi
Abstract
Each semester, students go through the process of selecting appropriate courses. It is difficult to find information about each course and ultimately make decisions. The objective of this paper is to design a course recommender model which takes student characteristics into account to recommend appropriate ...
Read More
Each semester, students go through the process of selecting appropriate courses. It is difficult to find information about each course and ultimately make decisions. The objective of this paper is to design a course recommender model which takes student characteristics into account to recommend appropriate courses. The model uses clustering to identify students with similar interests and skills. Once similar students are found, dependencies between student course selections are examined using fuzzy association rules mining. The application of clustering and fuzzy association rules results in appropriate recommendations and a predicted score. In this study, a collection of data on undergraduate students at the Management and Accounting Faculty of College of Farabi in University of Tehran is used. The records are from 2004 to 2015. The students are divided into two clusters according to Educational background and demographics. Finally, recommended courses and predicted scores are given to students. The mined rules facilitate decision-making regarding course selection.
Original/Review Paper
H.5. Image Processing and Computer Vision
S. Mavaddati
Abstract
In this paper, face detection problem is considered using the concepts of compressive sensing technique. This technique includes dictionary learning procedure and sparse coding method to represent the structural content of input images. In the proposed method, dictionaries are learned in such a way that ...
Read More
In this paper, face detection problem is considered using the concepts of compressive sensing technique. This technique includes dictionary learning procedure and sparse coding method to represent the structural content of input images. In the proposed method, dictionaries are learned in such a way that the trained models have the least degree of coherence to each other. The novelty of the proposed method involves the learning of comprehensive models with atoms that have the highest atom/data coherence with the training data and the lowest within-class and between-class coherence parameters. Each of these goals can be achieved through the proposed procedures. In order to achieve the desired results, a variety of features are extracted from the images and used to learn the characteristics of face and non-face images. Also, the results of the proposed classifier based on the incoherent dictionary learning technique are compared with the results obtained from the other common classifiers such as neural network and support vector machine. Simulation results, along with a significance statistical test show that the proposed method based on the incoherent models learned by the combinational features is able to detect the face regions with high accuracy rate.
Original/Review Paper
F.2.7. Optimization
F. Fouladi Mahani; A. Mahanipour; A. Mokhtari
Abstract
Recently, significant interest has been attracted by the potential use of aluminum nanostructures as plasmonic color filters to be great alternatives to the commercial color filters based on dye films or pigments. These color filters offer potential applications in LCDs, LEDs, color printing, CMOS image ...
Read More
Recently, significant interest has been attracted by the potential use of aluminum nanostructures as plasmonic color filters to be great alternatives to the commercial color filters based on dye films or pigments. These color filters offer potential applications in LCDs, LEDs, color printing, CMOS image sensors, and multispectral imaging. However, engineering the optical characteristics of these nanostructures to design a color filter with desired pass-band spectrum and high color purity requires accurate optimization techniques. In this paper, an optimization procedure integrating genetic algorithm with FDTD Solutions has been utilized to design plasmonic color filters, automatically. Our proposed aluminum nanohole arrays have been realized successfully to achieve additive (red, green, and blue) color filters using the automated optimization procedure. Despite all the considerations for fabrication simplicity, the designed filters feature transmission efficacies of 45-50 percent with a FWHM of 40 nm, 50 nm, and 80 nm for the red, green, and blue filters, respectively. The obtained results prove an efficient integration of genetic algorithm and FDTD Solutions revealing the potential application of the proposed method for automated design of similar nanostructures.
Review Article
H.5. Image Processing and Computer Vision
V. Patil; T. Sarode
Abstract
Image hashing allows compression, enhancement or other signal processing operations on digital images which are usually acceptable manipulations. Whereas, cryptographic hash functions are very sensitive to even single bit changes in image. Image hashing is a sum of important quality features in quantized ...
Read More
Image hashing allows compression, enhancement or other signal processing operations on digital images which are usually acceptable manipulations. Whereas, cryptographic hash functions are very sensitive to even single bit changes in image. Image hashing is a sum of important quality features in quantized form. In this paper, we proposed a novel image hashing algorithm for authentication which is more robust against various kind of attacks. In proposed approach, a short hash code is obtained by using minimum magnitude Center Symmetric Local Binary Pattern (CSLBP). The desirable discrimination power of image hash is maintained by modified Local Binary Pattern(LBP) based edge weight factor generated from gradient image. The proposed hashing method extracts texture features using the Center Symmetric Local Binary Pattern (CSLBP). The discrimination power of hashing is increased by weight factor during CSLBP histogram construction. The generated histogram is compressed to 1/4 of the original histogram by minimum magnitude CSLBP. The proposed method, has a twofold advantage, first is a small length and second is acceptable discrimination power. Experimental results are demonstrated by hamming distance, TPR, FPR and ROC curves. Therefore the proposed method successfully does a fair classification of content preserving and content changing images.
Original/Review Paper
H.6.2.5. Statistical
M. Mohammadi; M. Sarmad
Abstract
The purpose of this paper is to identify the effective points on the performance of one of the important algorithm of data mining namely support vector machine. The final classification decision has been made based on the small portion of data called support vectors. So, existence of the atypical observations ...
Read More
The purpose of this paper is to identify the effective points on the performance of one of the important algorithm of data mining namely support vector machine. The final classification decision has been made based on the small portion of data called support vectors. So, existence of the atypical observations in the aforementioned points, will result in deviation from the correct decision. Thus, the idea of Debruyne’s “outlier map” is employed in this paper to identify the outlying points in the SVM classification problem. However, due to the computational reasons such as convenience and rapidity, a robust Mahalanobis distance based on the minimum covariance determinant estimator is utilized. This method has a good compatibility by the data with low dimensional structure. In addition to the classification accuracy, the margin width is used as the criterion for the performance assessment. The larger margin is more desired, due to the higher generalization ability. It should be noted that, by omission of the detected outliers using the suggested outlier map the generalization ability and accuracy of SVM are increased. This leads to the conclusion that the proposed method is very efficient in identifying the outliers. The capability of recognizing the outlying and misclassified observations for this new version of outlier map has been retained similar to the older version, which is tested on the simulated and real world data.
Original/Review Paper
H.3.8. Natural Language Processing
A. Akkasi; E. Varoglu
Abstract
Chemical Named Entity Recognition (NER) is the basic step for consequent information extraction tasks such as named entity resolution, drug-drug interaction discovery, extraction of the names of the molecules and their properties. Improvement in the performance of such systems may affects the quality ...
Read More
Chemical Named Entity Recognition (NER) is the basic step for consequent information extraction tasks such as named entity resolution, drug-drug interaction discovery, extraction of the names of the molecules and their properties. Improvement in the performance of such systems may affects the quality of the subsequent tasks. Chemical text from which data for named entity recognition is extracted is naturally imbalanced since chemical entities are fewer compared to other segments in text. In this paper, the class imbalance problem in the context of chemical named entity recognition has been studied and adopted version of random undersampling for NER data, has been leveraged to generate a pool of classifiers. In order to keep the classes’ distribution balanced within each sentence, the well-known random undersampling method is modified to a sentence based version where the random removal of samples takes place within each sentence instead of considering the dataset as a whole. Furthermore, to take the advantages of combination of a set of diverse predictors, an ensemble of classifiers trained with the set of different training data resulted by sentence-based undersampling, is created. The proposed approach is developed and tested using the ChemDNER corpus released by BioCreative IV. Results show that the proposed method improves the classification performance of the baseline classifiers mainly as a result of an increase in recall. Furthermore, the combination of high performing classifiers trained using undersampled train data surpasses the performance of all single best classifiers and the combination of classifiers using full data.
Research Note
H.3.2.6. Games and infotainment
A.H. Khabbaz; A. Pouyan; M. Fateh; V. Abolghasemi
Abstract
This paper, presents an adapted serious game for rating social ability in children with autism spectrum disorder (ASD). The required measurements are obtained by challenges of the proposed serious game. The proposed serious game uses reinforcement learning concepts for being adaptive. It is based on ...
Read More
This paper, presents an adapted serious game for rating social ability in children with autism spectrum disorder (ASD). The required measurements are obtained by challenges of the proposed serious game. The proposed serious game uses reinforcement learning concepts for being adaptive. It is based on fuzzy logic to evaluate the social ability level of the children with ASD. The game adapts itself to the level of the autistic patient by reducing or increasing the challenges in the game via an intelligent agent during the play time. This task is accomplished by making more elements and reshaping them to a variety of real world shapes and redesigning their motions and speed. If autistic patient's communication level grows during the playtime, the challenges of game may become harder to make a dynamic procedure for evaluation. At each step or state, using fuzzy logic, the level of the player is estimated based on some attributes such as average of the distances between the fixed points gazed by the player, or number of the correct answers selected by the player divided by the number of the questioned objects. This paper offers the usage of dynamic AI difficulty system proposing a concept to enhance the conversation skills in autistic children. The proposed game is tested by participating of 3 autistic children. Each of them played the game in 5 turns. The results displays that the method is useful in the long-term.
Original/Review Paper
B.3. Communication/Networking and Information Technology
V. Babaiyan; Seyyede A. Sarfarazi
Abstract
Telecommunication Companies use data mining techniques to maintain good relationships with their existing customers and attract new customers and identifying profitable/unprofitable customers. Clustering leads to better understanding of customer and its results can be used to definition and decision-making ...
Read More
Telecommunication Companies use data mining techniques to maintain good relationships with their existing customers and attract new customers and identifying profitable/unprofitable customers. Clustering leads to better understanding of customer and its results can be used to definition and decision-making for promotional schemes. In this study, we used the 999-customer purchase records in South Khorasan Telecommunication Company which has been collected during a year. The purpose of this study is to classify customers into several clusters. Since the clusters and the number of their members are determined, high-consumption users will be logged out of the system and high-value customers who are missed will be identified. In this research we divided our customers into five categories: loyal, potential, new, missed and high-consumption by using the Clementine software, developing the RFM model to the LRFM model and TwoStep and k_Means algorithms. Thus, this category will be a good benchmark for company's future decisions and we can make better decisions for each group of customers in the future.
Original/Review Paper
M. Ilbeygi; M.R. Kangavari
Abstract
The increasing use of unmanned aerial vehicles (UAVs) or drones in different civil and military operations has attracted attention of many researchers and science communities. One of the most notable challenges in this field is supervising and controlling a group or a team of UAVs by a single user. Thereupon, ...
Read More
The increasing use of unmanned aerial vehicles (UAVs) or drones in different civil and military operations has attracted attention of many researchers and science communities. One of the most notable challenges in this field is supervising and controlling a group or a team of UAVs by a single user. Thereupon, we proposed a new intelligent adaptive interface (IAI) to overcome to this challenge. Our IAI not only is empowered by comprehensive IAI architecture but also has some notable features like: presenting single-display user interface for controlling UAV team, leveraging user cognitive model to deliver right information at the right time, supporting the user by system behavior explanation, guiding and helping the user to choose right decisions. Finally, we examined the IAI with contributing eleven volunteers and in three different scenarios. Obtained results have shown the power of the proposed IAI to reduce workload and to increase user's situation awareness level and as a result to promote mission completion percentage.
