Pub Date : 2020-11-01DOI: 10.1109/ICDMW51313.2020.00101
Jin-Young Kim, Sung-Bae Cho
Recently, deep learning models are utilized to predict the energy consumption. However, to construct the smart grid systems, the conventional methods have limitation on explanatory power or require manual analysis. To overcome it, in this paper, we present a novel deep learning model that can infer the predicted results by calculating the correlation between the latent variables and output as well as forecast the future consumption in high performance. The proposed model is composed of 1) a main encoder that models the past energy demand, 2) a sub encoder that models electric information except global active power as the latent variable in two dimensions, 3) a predictor that maps the future demand from the concatenation of the latent variables extracted from each encoder, and 4) an explainer that provides the most significant electric information. Several experiments on a household electric energy demand dataset show that the proposed model not only has better performance than the conventional models, but also provides the ability to explain the results by analyzing the correlation of inputs, latent variables, and energy demand predicted in the form of time-series.
{"title":"Electric Energy Demand Forecasting with Explainable Time-series Modeling","authors":"Jin-Young Kim, Sung-Bae Cho","doi":"10.1109/ICDMW51313.2020.00101","DOIUrl":"https://doi.org/10.1109/ICDMW51313.2020.00101","url":null,"abstract":"Recently, deep learning models are utilized to predict the energy consumption. However, to construct the smart grid systems, the conventional methods have limitation on explanatory power or require manual analysis. To overcome it, in this paper, we present a novel deep learning model that can infer the predicted results by calculating the correlation between the latent variables and output as well as forecast the future consumption in high performance. The proposed model is composed of 1) a main encoder that models the past energy demand, 2) a sub encoder that models electric information except global active power as the latent variable in two dimensions, 3) a predictor that maps the future demand from the concatenation of the latent variables extracted from each encoder, and 4) an explainer that provides the most significant electric information. Several experiments on a household electric energy demand dataset show that the proposed model not only has better performance than the conventional models, but also provides the ability to explain the results by analyzing the correlation of inputs, latent variables, and energy demand predicted in the form of time-series.","PeriodicalId":426846,"journal":{"name":"2020 International Conference on Data Mining Workshops (ICDMW)","volume":"349 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122041042","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 : 2020-11-01DOI: 10.1109/ICDMW51313.2020.00056
Ying Liu, Wei Wang, Tianlin Zhang, Zhenyu Cui
Learning effective feature interactions behind user behavior is challenging in credit scoring. Existing machine learning methods seem to have a strong bias towards low-order or high-order interactions, or require expertise feature engineering. In this paper, we present a novel neural network approach AttentionFM, which incorporates Factorization Machines and Attention mechanism for credit scoring. The proposed model focuses more on critical features and emphasizes both low- and high-order feature interactions, with no need of manually feature engineering on raw data representation. Experimental results demonstrate that our proposed model significantly outperforms the baselines based on two public datasets.
{"title":"AttentionFM: Incorporating Attention Mechanism and Factorization Machine for Credit Scoring","authors":"Ying Liu, Wei Wang, Tianlin Zhang, Zhenyu Cui","doi":"10.1109/ICDMW51313.2020.00056","DOIUrl":"https://doi.org/10.1109/ICDMW51313.2020.00056","url":null,"abstract":"Learning effective feature interactions behind user behavior is challenging in credit scoring. Existing machine learning methods seem to have a strong bias towards low-order or high-order interactions, or require expertise feature engineering. In this paper, we present a novel neural network approach AttentionFM, which incorporates Factorization Machines and Attention mechanism for credit scoring. The proposed model focuses more on critical features and emphasizes both low- and high-order feature interactions, with no need of manually feature engineering on raw data representation. Experimental results demonstrate that our proposed model significantly outperforms the baselines based on two public datasets.","PeriodicalId":426846,"journal":{"name":"2020 International Conference on Data Mining Workshops (ICDMW)","volume":"119 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117273944","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 : 2020-11-01DOI: 10.1109/ICDMW51313.2020.00115
Ritika Pandey, P. Brantingham, Craig D. Uchida, G. Mohler
Homicide investigations generate large and diverse data in the form of witness interview transcripts, physical evidence, photographs, DNA, etc. Homicide case chronologies are summaries of these data created by investigators that consist of short text-based entries documenting specific steps taken in the investigation. A chronology tracks the evolution of an investigation, including when and how persons involved and items of evidence became part of a case. In this article we discuss a framework for creating knowledge graphs of case chronologies that may aid investigators in analyzing homicide case data and also allow for post hoc analysis of the key features that determine whether a homicide is ultimately solved. Our method consists of 1) performing named entity recognition to determine witnesses, suspects, and detectives from chronology entries 2) using keyword expansion to identify documentary, physical, and forensic evidence in each entry and 3) linking entities and evidence to construct a homicide investigation knowledge graph. We compare the performance of several choices of methodologies for these sub-tasks using homicide investigation chronologies from Los Angeles, California. We then analyze the association between network statistics of the knowledge graphs and homicide solvability.
{"title":"Building knowledge graphs of homicide investigation chronologies","authors":"Ritika Pandey, P. Brantingham, Craig D. Uchida, G. Mohler","doi":"10.1109/ICDMW51313.2020.00115","DOIUrl":"https://doi.org/10.1109/ICDMW51313.2020.00115","url":null,"abstract":"Homicide investigations generate large and diverse data in the form of witness interview transcripts, physical evidence, photographs, DNA, etc. Homicide case chronologies are summaries of these data created by investigators that consist of short text-based entries documenting specific steps taken in the investigation. A chronology tracks the evolution of an investigation, including when and how persons involved and items of evidence became part of a case. In this article we discuss a framework for creating knowledge graphs of case chronologies that may aid investigators in analyzing homicide case data and also allow for post hoc analysis of the key features that determine whether a homicide is ultimately solved. Our method consists of 1) performing named entity recognition to determine witnesses, suspects, and detectives from chronology entries 2) using keyword expansion to identify documentary, physical, and forensic evidence in each entry and 3) linking entities and evidence to construct a homicide investigation knowledge graph. We compare the performance of several choices of methodologies for these sub-tasks using homicide investigation chronologies from Los Angeles, California. We then analyze the association between network statistics of the knowledge graphs and homicide solvability.","PeriodicalId":426846,"journal":{"name":"2020 International Conference on Data Mining Workshops (ICDMW)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123822929","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 : 2020-11-01DOI: 10.1109/ICDMW51313.2020.00128
Gejun Le, Qifeng Gu, Qingshan Jiang, Weiyi Lin
Supply chain involves mutual independent and distrusted stakeholders and large of sensitive order data. Sharing data among stakeholders is a essential project because that improves efficiency for various workflow among stakeholders. This paper proposes TrustedChain, a blockchain-based data sharing scheme for supply chain, which has two advantages: (a) trusted: we present a trusted environment, Trusted Environment (TE), based on blockchain to allow mutually distrusted stakeholders manage data collaboratively. (b) secure: we provide a secure design that first stores order forms in Distributed Database (DDB) and then records URI in Contract Account (CA) of TE. In addition, Supply-Business Contract Management (SCM) manages all CA and Node Communication (NC) allows communication over the network. The security analysis and evaluation prove the effectiveness of TrustedChain.
{"title":"TrustedChain: A Blockchain-based Data Sharing Scheme for Supply Chain","authors":"Gejun Le, Qifeng Gu, Qingshan Jiang, Weiyi Lin","doi":"10.1109/ICDMW51313.2020.00128","DOIUrl":"https://doi.org/10.1109/ICDMW51313.2020.00128","url":null,"abstract":"Supply chain involves mutual independent and distrusted stakeholders and large of sensitive order data. Sharing data among stakeholders is a essential project because that improves efficiency for various workflow among stakeholders. This paper proposes TrustedChain, a blockchain-based data sharing scheme for supply chain, which has two advantages: (a) trusted: we present a trusted environment, Trusted Environment (TE), based on blockchain to allow mutually distrusted stakeholders manage data collaboratively. (b) secure: we provide a secure design that first stores order forms in Distributed Database (DDB) and then records URI in Contract Account (CA) of TE. In addition, Supply-Business Contract Management (SCM) manages all CA and Node Communication (NC) allows communication over the network. The security analysis and evaluation prove the effectiveness of TrustedChain.","PeriodicalId":426846,"journal":{"name":"2020 International Conference on Data Mining Workshops (ICDMW)","volume":"10 1-2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134470362","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 : 2020-11-01DOI: 10.1109/ICDMW51313.2020.00087
A. T. Adebisi, V. Gonuguntla, Ho-Won Lee, K. Veluvolu
Dementia associated disorders such as vascular dementia, frontotemporal dementia and Alzheimer dementia lead to cognitive impairment. Discrimination of dementia associated disorders has reamined a challenging task as they have overlapping underlying complex structures and display similar clinical features. In this work, we explore an EEG based frequent subgraph searching technique to characterize stages of brain functional networks of mild cognitive impairment (MCI), Alzheimer's disease (AD) and vascular dementia (VD) subjects in comparison with healthy control (HC) subjects. To identify the frequent subgraph related to dementia, we first formulated the brain functional network based on the phase information of EEG with mutual information as a measure. The whole network is then divided into sub-regions and frequent sub-graph search is performed. The identified frequent subgraphs were employed to discriminate the dementia associated disorders from the data recorded from 10 healthy and 32 dementia subjects in various stages. Results show that the proposed method has the potential to quantify the disease progression using brain functional connectivity and the identified networks can aid in the diagnosis of dementia associated disorders.
{"title":"Classification of Dementia Associated Disorders Using EEG based Frequent Subgraph Technique","authors":"A. T. Adebisi, V. Gonuguntla, Ho-Won Lee, K. Veluvolu","doi":"10.1109/ICDMW51313.2020.00087","DOIUrl":"https://doi.org/10.1109/ICDMW51313.2020.00087","url":null,"abstract":"Dementia associated disorders such as vascular dementia, frontotemporal dementia and Alzheimer dementia lead to cognitive impairment. Discrimination of dementia associated disorders has reamined a challenging task as they have overlapping underlying complex structures and display similar clinical features. In this work, we explore an EEG based frequent subgraph searching technique to characterize stages of brain functional networks of mild cognitive impairment (MCI), Alzheimer's disease (AD) and vascular dementia (VD) subjects in comparison with healthy control (HC) subjects. To identify the frequent subgraph related to dementia, we first formulated the brain functional network based on the phase information of EEG with mutual information as a measure. The whole network is then divided into sub-regions and frequent sub-graph search is performed. The identified frequent subgraphs were employed to discriminate the dementia associated disorders from the data recorded from 10 healthy and 32 dementia subjects in various stages. Results show that the proposed method has the potential to quantify the disease progression using brain functional connectivity and the identified networks can aid in the diagnosis of dementia associated disorders.","PeriodicalId":426846,"journal":{"name":"2020 International Conference on Data Mining Workshops (ICDMW)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133765864","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 : 2020-11-01DOI: 10.1109/ICDMW51313.2020.00110
Hyung-Jun Moon, Seok-Jun Bu, Sung-Bae Cho
In the time-series models for predicting residential energy consumption, the energy properties collected through multiple sensors usually include irregular and seasonal factors. The irregular pattern resulting from them is called peak demand, which is a major cause of performance degradation. In order to enhance the performance, we propose a convolutional-recurrent triplet network to learn and detect the demand peaks. The proposed model generates the latent space for demand peaks from data, which is transferred into convolutional neural network-long short-term memory (CNN-LSTM) to finally predict the future power demand. Experiments with the dataset of UCI household power consumption composed of a total of 2,075,259 time-series data show that the proposed model reduces the error by 23.63% and outperforms the state-of-the-art deep learning models including the CNN-LSTM. Especially, the proposed model improves the prediction performance by modeling the distribution of demand peaks in Euclidean space.
{"title":"Learning Disentangled Representation of Residential Power Demand Peak via Convolutional-Recurrent Triplet Network","authors":"Hyung-Jun Moon, Seok-Jun Bu, Sung-Bae Cho","doi":"10.1109/ICDMW51313.2020.00110","DOIUrl":"https://doi.org/10.1109/ICDMW51313.2020.00110","url":null,"abstract":"In the time-series models for predicting residential energy consumption, the energy properties collected through multiple sensors usually include irregular and seasonal factors. The irregular pattern resulting from them is called peak demand, which is a major cause of performance degradation. In order to enhance the performance, we propose a convolutional-recurrent triplet network to learn and detect the demand peaks. The proposed model generates the latent space for demand peaks from data, which is transferred into convolutional neural network-long short-term memory (CNN-LSTM) to finally predict the future power demand. Experiments with the dataset of UCI household power consumption composed of a total of 2,075,259 time-series data show that the proposed model reduces the error by 23.63% and outperforms the state-of-the-art deep learning models including the CNN-LSTM. Especially, the proposed model improves the prediction performance by modeling the distribution of demand peaks in Euclidean space.","PeriodicalId":426846,"journal":{"name":"2020 International Conference on Data Mining Workshops (ICDMW)","volume":"150 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122762712","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 : 2020-11-01DOI: 10.1109/ICDMW51313.2020.00109
Sue Hyang Lim, S. Kim, Hyeong Min Lee, Sijun Kim, Y. Shin
Rapid charging of Li-ion batteries is vital for the commercialization of electric propulsion systems. But, during the fast-charging process, reduction in the battery capacity and temperature increases must be considered in real-time. Most Li-ion battery chargers follow the charging profile of an open-loop system, which has been determined based on prior knowledge. However, such a system does not reflect the temperature change of the battery and the degree of aging. Therefore, in this study, we propose a neural network-based charging profile model by applying a closed-loop system to reflect the various states of batteries; we also show two battery-state characteristics in addition to temperature. Consequently, we show battery characteristics other than those shown in the past, such as the battery voltage and temperature trends. In addition to the design of the charging current, an improvement of approximately 22 ∼ 50% based on the mean absolute error (MAE) is achieved. By considering the various characteristics, the long short-term memory performance is determined to be better when compared to the feed-forward neural network, and this performance is improved by 35% based on MAE.
{"title":"Design of Neural Network-based Boost Charging for Reducing the Charging Time of Li-ion Battery","authors":"Sue Hyang Lim, S. Kim, Hyeong Min Lee, Sijun Kim, Y. Shin","doi":"10.1109/ICDMW51313.2020.00109","DOIUrl":"https://doi.org/10.1109/ICDMW51313.2020.00109","url":null,"abstract":"Rapid charging of Li-ion batteries is vital for the commercialization of electric propulsion systems. But, during the fast-charging process, reduction in the battery capacity and temperature increases must be considered in real-time. Most Li-ion battery chargers follow the charging profile of an open-loop system, which has been determined based on prior knowledge. However, such a system does not reflect the temperature change of the battery and the degree of aging. Therefore, in this study, we propose a neural network-based charging profile model by applying a closed-loop system to reflect the various states of batteries; we also show two battery-state characteristics in addition to temperature. Consequently, we show battery characteristics other than those shown in the past, such as the battery voltage and temperature trends. In addition to the design of the charging current, an improvement of approximately 22 ∼ 50% based on the mean absolute error (MAE) is achieved. By considering the various characteristics, the long short-term memory performance is determined to be better when compared to the feed-forward neural network, and this performance is improved by 35% based on MAE.","PeriodicalId":426846,"journal":{"name":"2020 International Conference on Data Mining Workshops (ICDMW)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123358265","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 : 2020-11-01DOI: 10.1109/ICDMW51313.2020.00073
Syed Hasan Amin Mahmood, A. Abbasi
Time series predictions are important for various application domains. However, effective forecasting can be challenging in noisy contexts devoid of time series data encompassing stationarity, cyclicality, completeness, and non-sparseness. Cyber-security is a good example of such context. In organizational security settings, predicting time series related to emerging attacks could enhance cyber threat intelligence, resulting in timely and actionable insights at the operational, tactical, and strategic levels. In order to explore this gap, we propose a deep generative model-based framework for time series forecasting in noisy data environments. The proposed framework incorporates a novel ensembling strategy where generative adversarial networks and recurrent variational autoencoders are leveraged in unison with base predictors for enhanced regularization of time series predictive models. The framework is extensible, supporting different model combinations and analytical or iterative model fusion strategies. Using a test bed encompassing 10 years of weekly phishing attack volume data from 5 organizations in the technology, financial services, and social networking sectors, we show that the framework can boost predictive power for various standard time series models. Additional results reveal that the framework outperforms generative data augmentation approaches designed to enrich the input time series data matrices. Collectively, our findings suggest that utilizing generative models in more robust end-to-end setup can improve prediction in cyber threat intelligence contexts, as well as related problems involving challenging time series data.
{"title":"Using Deep Generative Models to Boost Forecasting: A Phishing Prediction Case Study","authors":"Syed Hasan Amin Mahmood, A. Abbasi","doi":"10.1109/ICDMW51313.2020.00073","DOIUrl":"https://doi.org/10.1109/ICDMW51313.2020.00073","url":null,"abstract":"Time series predictions are important for various application domains. However, effective forecasting can be challenging in noisy contexts devoid of time series data encompassing stationarity, cyclicality, completeness, and non-sparseness. Cyber-security is a good example of such context. In organizational security settings, predicting time series related to emerging attacks could enhance cyber threat intelligence, resulting in timely and actionable insights at the operational, tactical, and strategic levels. In order to explore this gap, we propose a deep generative model-based framework for time series forecasting in noisy data environments. The proposed framework incorporates a novel ensembling strategy where generative adversarial networks and recurrent variational autoencoders are leveraged in unison with base predictors for enhanced regularization of time series predictive models. The framework is extensible, supporting different model combinations and analytical or iterative model fusion strategies. Using a test bed encompassing 10 years of weekly phishing attack volume data from 5 organizations in the technology, financial services, and social networking sectors, we show that the framework can boost predictive power for various standard time series models. Additional results reveal that the framework outperforms generative data augmentation approaches designed to enrich the input time series data matrices. Collectively, our findings suggest that utilizing generative models in more robust end-to-end setup can improve prediction in cyber threat intelligence contexts, as well as related problems involving challenging time series data.","PeriodicalId":426846,"journal":{"name":"2020 International Conference on Data Mining Workshops (ICDMW)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115744481","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 : 2020-11-01DOI: 10.1109/ICDMW51313.2020.00064
Christian Schreckenberger, Tim Glockner, H. Stuckenschmidt, Christian Bartelt
Trapezoidal Data Streams are an emerging topic, where not only the data volume increases, but also the data dimension, i.e. new features emerge. In this paper, we address the challenges that arise from this problem by providing a novel approach to restructure and prune Hoeffding trees. We evaluate our approach on synthetic datasets, where we can show that the approach significantly improves the performance compared to the baseline of an adjusted Hoeffding tree algorithm without restructuring and pruning.
{"title":"Restructuring of Hoeffding Trees for Trapezoidal Data Streams","authors":"Christian Schreckenberger, Tim Glockner, H. Stuckenschmidt, Christian Bartelt","doi":"10.1109/ICDMW51313.2020.00064","DOIUrl":"https://doi.org/10.1109/ICDMW51313.2020.00064","url":null,"abstract":"Trapezoidal Data Streams are an emerging topic, where not only the data volume increases, but also the data dimension, i.e. new features emerge. In this paper, we address the challenges that arise from this problem by providing a novel approach to restructure and prune Hoeffding trees. We evaluate our approach on synthetic datasets, where we can show that the approach significantly improves the performance compared to the baseline of an adjusted Hoeffding tree algorithm without restructuring and pruning.","PeriodicalId":426846,"journal":{"name":"2020 International Conference on Data Mining Workshops (ICDMW)","volume":"167 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114661349","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 : 2020-11-01DOI: 10.1109/ICDMW51313.2020.00097
J. Wu, Qian Teng, Gautam Srivastava, Matin Pirouz, Chun-Wei Lin
In this paper, we propose a new pattern called skyline quantity-utility pattern (SQUP) to provide better estimations in the decision-making process by considering quantity and utility together. Two algorithms respectively called SQUM-1 and SQUM-2 are presented to efficiently mine the set of SQUPs. Two new efficient utility-max structures are also mentioned for the reduction of the candidate itemsets respectively utilized in two developed algorithms. Our in-depth experimental results prove that our proposed algorithms achieve good performance in terms of runtime and memory usage.
{"title":"Efficient Mining of Non-Dominated High Quantity-Utility Patterns","authors":"J. Wu, Qian Teng, Gautam Srivastava, Matin Pirouz, Chun-Wei Lin","doi":"10.1109/ICDMW51313.2020.00097","DOIUrl":"https://doi.org/10.1109/ICDMW51313.2020.00097","url":null,"abstract":"In this paper, we propose a new pattern called skyline quantity-utility pattern (SQUP) to provide better estimations in the decision-making process by considering quantity and utility together. Two algorithms respectively called SQUM-1 and SQUM-2 are presented to efficiently mine the set of SQUPs. Two new efficient utility-max structures are also mentioned for the reduction of the candidate itemsets respectively utilized in two developed algorithms. Our in-depth experimental results prove that our proposed algorithms achieve good performance in terms of runtime and memory usage.","PeriodicalId":426846,"journal":{"name":"2020 International Conference on Data Mining Workshops (ICDMW)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123810350","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：