Pub Date : 2015-12-01DOI: 10.1109/SPIS.2015.7422307
Maryam Dialameh, M. Z. Jahromi
Most of data mining algorithms including classifiers suffer from data sets with highly imbalanced distribution of the target variable. The problem becomes more serious when the events have different costs. Feature weighting and instance weighting are two most common ways to tackle this problem. However, none of the current weighting methods take into account the salience of features. In order to accomplish this, a novel and flexible weighting function is proposed that dynamically assigns a proper weight to each feature. Experiments results show that the proposed weighting function is superior to current methods.
{"title":"Dynamic feature weighting for imbalanced data sets","authors":"Maryam Dialameh, M. Z. Jahromi","doi":"10.1109/SPIS.2015.7422307","DOIUrl":"https://doi.org/10.1109/SPIS.2015.7422307","url":null,"abstract":"Most of data mining algorithms including classifiers suffer from data sets with highly imbalanced distribution of the target variable. The problem becomes more serious when the events have different costs. Feature weighting and instance weighting are two most common ways to tackle this problem. However, none of the current weighting methods take into account the salience of features. In order to accomplish this, a novel and flexible weighting function is proposed that dynamically assigns a proper weight to each feature. Experiments results show that the proposed weighting function is superior to current methods.","PeriodicalId":424434,"journal":{"name":"2015 Signal Processing and Intelligent Systems Conference (SPIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127459612","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}
Pub Date : 2015-12-01DOI: 10.1109/SPIS.2015.7422303
Parisa Karimi, F. Farzaneh
The multi-path phenomenon is one of the main causes of error in Direction Of Arrival Estimation systems which are located in a complex environment. The received signal in this environment consists of Line Of Sight (LOS) and multi-path components which are delayed versions of the LOS signal with the amplitude and the phase depending on the path length. In order to eliminate the multi-path signal, it is necessary to estimate the amplitude, the phase, and the delay of the signal. To this end, a method based on correlation which has been already used to estimate amplitudes and delays of Line Of Sight and multi-path signals in GPS systems, is implemented. After this estimation process, a phase estimation algorithm is proposed in order to construct the multi-path signal replica adequately. Finally, by subtracting the estimated multi-path component from the total signal, the LOS component could be extracted. Simulations show that this algorithm eliminates adequately the multi-path component and results in a better DOA estimation.
{"title":"Reduction of multi-path effect based on correlation decomposition in a DOA estimation system","authors":"Parisa Karimi, F. Farzaneh","doi":"10.1109/SPIS.2015.7422303","DOIUrl":"https://doi.org/10.1109/SPIS.2015.7422303","url":null,"abstract":"The multi-path phenomenon is one of the main causes of error in Direction Of Arrival Estimation systems which are located in a complex environment. The received signal in this environment consists of Line Of Sight (LOS) and multi-path components which are delayed versions of the LOS signal with the amplitude and the phase depending on the path length. In order to eliminate the multi-path signal, it is necessary to estimate the amplitude, the phase, and the delay of the signal. To this end, a method based on correlation which has been already used to estimate amplitudes and delays of Line Of Sight and multi-path signals in GPS systems, is implemented. After this estimation process, a phase estimation algorithm is proposed in order to construct the multi-path signal replica adequately. Finally, by subtracting the estimated multi-path component from the total signal, the LOS component could be extracted. Simulations show that this algorithm eliminates adequately the multi-path component and results in a better DOA estimation.","PeriodicalId":424434,"journal":{"name":"2015 Signal Processing and Intelligent Systems Conference (SPIS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122102536","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}
Pub Date : 2015-12-01DOI: 10.1109/SPIS.2015.7422302
M. Rad, A. Shahbahrami
Tax fraud includes a large spectrum of methods to deny the facts and realities, claiming wrong information, and accomplishing financial businesses regardless of what the legal frameworks are. Nowadays, with the development tax systems and the large volume of the data stored in them, need is felt for a tool by which we can process the stored data and provide users with the information obtained from it. According to tax politics, especially value-added tax, the rate of tax fraud is now increasing. Based on the investigations, recent researchers tend to use similar and standard methods to detect tax fraud, which includes, association rules, clustering, neural networks, decision trees, Bayesian networks, regression and genetic algorithms. Because of large volume of tax database, most of the studied methods about fraud detection are computationally intensive. In order to increase the performance of fraud detection algorithms such as Bayesian networks, parallelism techniques are used in this paper. We used parallel technology of Microsoft .Net, parallel loops and P-LINQ on the Intel Xeon server with 16, X7755 dual core processors and memory of 32GB. The implementation results on real database show that a speedup of up to 9.2x is achieved.
{"title":"High performance implementation of tax fraud detection algorithm","authors":"M. Rad, A. Shahbahrami","doi":"10.1109/SPIS.2015.7422302","DOIUrl":"https://doi.org/10.1109/SPIS.2015.7422302","url":null,"abstract":"Tax fraud includes a large spectrum of methods to deny the facts and realities, claiming wrong information, and accomplishing financial businesses regardless of what the legal frameworks are. Nowadays, with the development tax systems and the large volume of the data stored in them, need is felt for a tool by which we can process the stored data and provide users with the information obtained from it. According to tax politics, especially value-added tax, the rate of tax fraud is now increasing. Based on the investigations, recent researchers tend to use similar and standard methods to detect tax fraud, which includes, association rules, clustering, neural networks, decision trees, Bayesian networks, regression and genetic algorithms. Because of large volume of tax database, most of the studied methods about fraud detection are computationally intensive. In order to increase the performance of fraud detection algorithms such as Bayesian networks, parallelism techniques are used in this paper. We used parallel technology of Microsoft .Net, parallel loops and P-LINQ on the Intel Xeon server with 16, X7755 dual core processors and memory of 32GB. The implementation results on real database show that a speedup of up to 9.2x is achieved.","PeriodicalId":424434,"journal":{"name":"2015 Signal Processing and Intelligent Systems Conference (SPIS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127784057","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}
Pub Date : 2015-12-01DOI: 10.1109/SPIS.2015.7422316
Masoumeh Rezaei Abkenar, M. Ahmad
A salient region is the most distinctive part of the image that captures human's attention. Saliency detection is a fundamental characteristic of the human visual system. Finding computational models which are able to detect salient regions is a challenging task for image processing and computer vision applications. Salient regions of various sizes can be detected from different scales. Therefore, selecting the best scales is an important issue. In this paper, an efficient multi-scale method to find salient regions is proposed. In order to include more features in evaluating saliency of a pixel, feature maps are generated using components of both the RGB and YUV color spaces. These features are combined into quaternions. Detecting salient regions of different sizes is addressed by utilizing a scale space analysis. Salient regions are detected by convolving the image amplitude spectrum with a low-pass Gaussian kernel of multiple scales. To incorporate more meaningful information, more than one scale is considered based on entropy criterion. The final saliency map is generated by normalizing the weighted saliency maps of these scales. Experiments are conducted on a dataset of natural images to evaluate the performance of the proposed method. Results show that the proposed method provides larger values of area under receiver operating characteristics curve, precision, recall and F-measure, in comparison to some of the state-of-the-art methods.
{"title":"Quaternion-based salient region detection using scale space analysis","authors":"Masoumeh Rezaei Abkenar, M. Ahmad","doi":"10.1109/SPIS.2015.7422316","DOIUrl":"https://doi.org/10.1109/SPIS.2015.7422316","url":null,"abstract":"A salient region is the most distinctive part of the image that captures human's attention. Saliency detection is a fundamental characteristic of the human visual system. Finding computational models which are able to detect salient regions is a challenging task for image processing and computer vision applications. Salient regions of various sizes can be detected from different scales. Therefore, selecting the best scales is an important issue. In this paper, an efficient multi-scale method to find salient regions is proposed. In order to include more features in evaluating saliency of a pixel, feature maps are generated using components of both the RGB and YUV color spaces. These features are combined into quaternions. Detecting salient regions of different sizes is addressed by utilizing a scale space analysis. Salient regions are detected by convolving the image amplitude spectrum with a low-pass Gaussian kernel of multiple scales. To incorporate more meaningful information, more than one scale is considered based on entropy criterion. The final saliency map is generated by normalizing the weighted saliency maps of these scales. Experiments are conducted on a dataset of natural images to evaluate the performance of the proposed method. Results show that the proposed method provides larger values of area under receiver operating characteristics curve, precision, recall and F-measure, in comparison to some of the state-of-the-art methods.","PeriodicalId":424434,"journal":{"name":"2015 Signal Processing and Intelligent Systems Conference (SPIS)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127065922","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}
Pub Date : 2015-12-01DOI: 10.1109/SPIS.2015.7422326
F. Jafarinejad, A. Pouyan
Computerized Intelligent manufacturing systems are a well-known example of discrete event systems. Concurrency and asynchronous nature of these systems imply that solutions of classical control can't be effective in this domain. Moreover, in real world manufacturing systems such as cable manufacturing systems design and analysis of the large scale system awards a new problem in automated design and diagnosis. Petri net as a high level graphical and formal specification language has the ability of modeling most of properties of these systems such as non-determinism, concurrency, mutual exclusion etc. They offer a solution both in modeling and verification of these systems. Furthermore, different Petri net transformation techniques grant a divide and conquer approach for large scale real world systems. This paper offers a modular hybrid model for designing a cable manufacturing systems and its verification. Giving well-behaved modules of system, this approach outputs a well-behaved system through some property preserving transformations without extra verification of the whole system. Finally, the model is able to be compiled into control codes and implemented in hardware to control procedure of system runs.
{"title":"A modular synthesis approach for intelligent manufacturing system design: A Petri net based transformation method","authors":"F. Jafarinejad, A. Pouyan","doi":"10.1109/SPIS.2015.7422326","DOIUrl":"https://doi.org/10.1109/SPIS.2015.7422326","url":null,"abstract":"Computerized Intelligent manufacturing systems are a well-known example of discrete event systems. Concurrency and asynchronous nature of these systems imply that solutions of classical control can't be effective in this domain. Moreover, in real world manufacturing systems such as cable manufacturing systems design and analysis of the large scale system awards a new problem in automated design and diagnosis. Petri net as a high level graphical and formal specification language has the ability of modeling most of properties of these systems such as non-determinism, concurrency, mutual exclusion etc. They offer a solution both in modeling and verification of these systems. Furthermore, different Petri net transformation techniques grant a divide and conquer approach for large scale real world systems. This paper offers a modular hybrid model for designing a cable manufacturing systems and its verification. Giving well-behaved modules of system, this approach outputs a well-behaved system through some property preserving transformations without extra verification of the whole system. Finally, the model is able to be compiled into control codes and implemented in hardware to control procedure of system runs.","PeriodicalId":424434,"journal":{"name":"2015 Signal Processing and Intelligent Systems Conference (SPIS)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128191718","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}
Pub Date : 2015-12-01DOI: 10.1109/SPIS.2015.7422333
M. Jamali, Vahid Ghafarinia, M. A. Montazeri
Vector quantization (VQ) is a fast and simple classification algorithm that has been widely employed for the recognition of isolated spoken words. However, this algorithm and most of its improved versions fail to accurately distinguish words with similar vowels. The spoken pattern of digits/dow/ and/noh/ (2 and 9 respectively) in Persian is a good example for this type of similarity. In this paper we have proposed an enhanced vector quantization algorithm in which the deterministic role of the short consonants at the beginning of the words is taken into account. In this algorithm an unknown vector is judged based on the classification results of two set of codebooks. The first set of codebooks is constructed by the initial portions of the words while the other set is constructed by the whole words. The performance of the proposed algorithm was experimentally verified against other VQ-based algorithms. While the overall performance of the proposed algorithm was above the others, in the case of similar words it could remarkably decrease the number of misclassification. This improvement was achieved by only a small increase in the computational load.
{"title":"Recognition of speaker-independent isolated Persian digits using an enhanced vector quantization algorithm","authors":"M. Jamali, Vahid Ghafarinia, M. A. Montazeri","doi":"10.1109/SPIS.2015.7422333","DOIUrl":"https://doi.org/10.1109/SPIS.2015.7422333","url":null,"abstract":"Vector quantization (VQ) is a fast and simple classification algorithm that has been widely employed for the recognition of isolated spoken words. However, this algorithm and most of its improved versions fail to accurately distinguish words with similar vowels. The spoken pattern of digits/dow/ and/noh/ (2 and 9 respectively) in Persian is a good example for this type of similarity. In this paper we have proposed an enhanced vector quantization algorithm in which the deterministic role of the short consonants at the beginning of the words is taken into account. In this algorithm an unknown vector is judged based on the classification results of two set of codebooks. The first set of codebooks is constructed by the initial portions of the words while the other set is constructed by the whole words. The performance of the proposed algorithm was experimentally verified against other VQ-based algorithms. While the overall performance of the proposed algorithm was above the others, in the case of similar words it could remarkably decrease the number of misclassification. This improvement was achieved by only a small increase in the computational load.","PeriodicalId":424434,"journal":{"name":"2015 Signal Processing and Intelligent Systems Conference (SPIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128238374","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}
Pub Date : 2015-12-01DOI: 10.1109/SPIS.2015.7422313
M. Taghizadeh, A. Chalechale
Multiple-query image retrieval is usually utilized in order to enhance performance of the image retrieval system with considering single semantic for a query set. So far, multiple-query image retrieval based on different queries has rarely studied. In this work, we intend to address this problem using a binary component vector. This vector indicates distinct components which exist in an image. The binary component vector is also generated utilizing low-level feature extraction techniques. The final image retrieval process is performed based on this vector. The experimental results show a better performance and less computation in contrary to previous proposed methods.
{"title":"A novel method for multiple-query image retrieval","authors":"M. Taghizadeh, A. Chalechale","doi":"10.1109/SPIS.2015.7422313","DOIUrl":"https://doi.org/10.1109/SPIS.2015.7422313","url":null,"abstract":"Multiple-query image retrieval is usually utilized in order to enhance performance of the image retrieval system with considering single semantic for a query set. So far, multiple-query image retrieval based on different queries has rarely studied. In this work, we intend to address this problem using a binary component vector. This vector indicates distinct components which exist in an image. The binary component vector is also generated utilizing low-level feature extraction techniques. The final image retrieval process is performed based on this vector. The experimental results show a better performance and less computation in contrary to previous proposed methods.","PeriodicalId":424434,"journal":{"name":"2015 Signal Processing and Intelligent Systems Conference (SPIS)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131258274","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}
Pub Date : 2015-12-01DOI: 10.1109/SPIS.2015.7422324
Marziyeh Meamaripour, Mohammad Saberali
This paper deals with the relationship of the Cramer-Rao lower bound (CRLB) and the channel capacity in a channel with interference. The former quantity is the representative of estimation theory and the latter is the envoy to information theory. CRLB is the lower bound on the variance of any unbiased estimator and channel capacity is the upper bound on the rate of reliable transmission. In the region that signal-to-interference ratio (SIR) is close to zero, the local minimum of the capacity of binary channel and local maximum of the CRLB happens. The log-likelihood ratio (LLR) is investigated to interpret the estimation and the capacity characteristics and plays a main role in this way.
{"title":"On the relationship of the Cramer-Rao lower bound and channel capacity in an interfered binary channel through the log-likelihood ratio","authors":"Marziyeh Meamaripour, Mohammad Saberali","doi":"10.1109/SPIS.2015.7422324","DOIUrl":"https://doi.org/10.1109/SPIS.2015.7422324","url":null,"abstract":"This paper deals with the relationship of the Cramer-Rao lower bound (CRLB) and the channel capacity in a channel with interference. The former quantity is the representative of estimation theory and the latter is the envoy to information theory. CRLB is the lower bound on the variance of any unbiased estimator and channel capacity is the upper bound on the rate of reliable transmission. In the region that signal-to-interference ratio (SIR) is close to zero, the local minimum of the capacity of binary channel and local maximum of the CRLB happens. The log-likelihood ratio (LLR) is investigated to interpret the estimation and the capacity characteristics and plays a main role in this way.","PeriodicalId":424434,"journal":{"name":"2015 Signal Processing and Intelligent Systems Conference (SPIS)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132993592","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}
Pub Date : 2015-12-01DOI: 10.1109/SPIS.2015.7422328
Arezoo Mousavi, S. S. Ghidary, Zohre Karimi
Network security has become one of the well-known concerns in the last decades. Machine learning techniques are robust methods in detecting malicious activities and network threats. Most previous works learn offline supervised classifiers while they require large amounts of labeled examples and also should update models because the data change over time in real world applications. To alleviate these problems, we propose a novel online version of laplacian twin support vector machine classifier, which can exploit the geometry information of the marginal distribution embedded in unlabeled data to construct a more accurate and faster semi-supervised classifier. The results of experiments on large network datasets show that Online Lap-TSVM combined by two nonparallel hyper planes improves the accuracy with the comparable computing time and storage to Lap-TSVM.
{"title":"Semi-supervised intrusion detection via online laplacian twin support vector machine","authors":"Arezoo Mousavi, S. S. Ghidary, Zohre Karimi","doi":"10.1109/SPIS.2015.7422328","DOIUrl":"https://doi.org/10.1109/SPIS.2015.7422328","url":null,"abstract":"Network security has become one of the well-known concerns in the last decades. Machine learning techniques are robust methods in detecting malicious activities and network threats. Most previous works learn offline supervised classifiers while they require large amounts of labeled examples and also should update models because the data change over time in real world applications. To alleviate these problems, we propose a novel online version of laplacian twin support vector machine classifier, which can exploit the geometry information of the marginal distribution embedded in unlabeled data to construct a more accurate and faster semi-supervised classifier. The results of experiments on large network datasets show that Online Lap-TSVM combined by two nonparallel hyper planes improves the accuracy with the comparable computing time and storage to Lap-TSVM.","PeriodicalId":424434,"journal":{"name":"2015 Signal Processing and Intelligent Systems Conference (SPIS)","volume":"4 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129088485","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}
Pub Date : 2015-12-01DOI: 10.1109/SPIS.2015.7422311
Fatemeh Eskandari, Hamid Shayestehmanesh, S. Hashemi
While interests in seeking and sharing questions/ answers through the Community Question Answering (CQA) systems has been increased, predicting the best answer in such systems is one of the main challenges that we are going to tackle in this paper. Considering comments as one of the inputs in our model and extracting features using Natural Language Processing (NLP) and text mining techniques such as Sentiment Analysis (SA) on comments and spell checking for answers, are the main parts of this research. Moreover, we worked on English language websites. On the other hand, users' social behavior and their activities considered as informative features in this paper. As a result, by finding the best combination of different features the performance of our model shows improvement in comparison to the related previous works on "Stack Exchange" websites.
{"title":"Predicting best answer using sentiment analysis in community question answering systems","authors":"Fatemeh Eskandari, Hamid Shayestehmanesh, S. Hashemi","doi":"10.1109/SPIS.2015.7422311","DOIUrl":"https://doi.org/10.1109/SPIS.2015.7422311","url":null,"abstract":"While interests in seeking and sharing questions/ answers through the Community Question Answering (CQA) systems has been increased, predicting the best answer in such systems is one of the main challenges that we are going to tackle in this paper. Considering comments as one of the inputs in our model and extracting features using Natural Language Processing (NLP) and text mining techniques such as Sentiment Analysis (SA) on comments and spell checking for answers, are the main parts of this research. Moreover, we worked on English language websites. On the other hand, users' social behavior and their activities considered as informative features in this paper. As a result, by finding the best combination of different features the performance of our model shows improvement in comparison to the related previous works on \"Stack Exchange\" websites.","PeriodicalId":424434,"journal":{"name":"2015 Signal Processing and Intelligent Systems Conference (SPIS)","volume":"359 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132619307","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：