To an extent as never before in the history of medicine, computers are supporting human input, decision making and provision of data. In today’s healthcare sector and medical profession, AI, algorithms, robotics and big data are used to derive inferences for monitoring large-scale medical trends, detecting and measuring individual risks and chances based on data-driven estimations. A knowledge-intensive industry like the healthcare profession highly depends on data and analytics to improve therapies and practices. In recent years, there has been tremendous growth in the range of medical information collected, including clinical, genetic, behavioral and environmental data. Every day, healthcare professionals, biomedical researchers and patients produce vast amounts of data from an array of devices. These include electronic health records (EHRs), genome sequencing machines, high-resolution medical imaging, smartphone applications and ubiquitous sensing, as well as Internet of Things (IoT) devices that monitor patient health (OECD 2015). Through machine learning algorithms and unprecedented data storage and computational power, AI technologies have most advanced abilities to gain information, process it and give a well-defined output to the end-user. Daily monitoring thereby aids to create big data to recognize behavioral patterns’ relation to health status in order to create predictions with highest mathematical precision based on big data capturing large-scale samples. AI thereby enlightens to analyze the relation between prevention and treatment and patient outcomes in all stages of diagnosis, treatment, drug development and monitoring, personalized medicine, patient control and care. Advanced hospitals are looking into AI solutions to support and perform operational initiatives that increase precision and cost effectiveness. Robotics have been used for disabled and patient care assistance. Medical decision making has been supported through predictive analytics and general healthcare management technology. Network connectivity allows access to affordable healthcare around the globe in a cost-effective way.
{"title":"Artificial Intelligence in the Healthcare Sector","authors":"Julia M. Puaschunder, Dieter Feierabend","doi":"10.2139/ssrn.3469423","DOIUrl":"https://doi.org/10.2139/ssrn.3469423","url":null,"abstract":"To an extent as never before in the history of medicine, computers are supporting human input, decision making and provision of data. In today’s healthcare sector and medical profession, AI, algorithms, robotics and big data are used to derive inferences for monitoring large-scale medical trends, detecting and measuring individual risks and chances based on data-driven estimations. A knowledge-intensive industry like the healthcare profession highly depends on data and analytics to improve therapies and practices. In recent years, there has been tremendous growth in the range of medical information collected, including clinical, genetic, behavioral and environmental data. Every day, healthcare professionals, biomedical researchers and patients produce vast amounts of data from an array of devices. These include electronic health records (EHRs), genome sequencing machines, high-resolution medical imaging, smartphone applications and ubiquitous sensing, as well as Internet of Things (IoT) devices that monitor patient health (OECD 2015). Through machine learning algorithms and unprecedented data storage and computational power, AI technologies have most advanced abilities to gain information, process it and give a well-defined output to the end-user. Daily monitoring thereby aids to create big data to recognize behavioral patterns’ relation to health status in order to create predictions with highest mathematical precision based on big data capturing large-scale samples. AI thereby enlightens to analyze the relation between prevention and treatment and patient outcomes in all stages of diagnosis, treatment, drug development and monitoring, personalized medicine, patient control and care. Advanced hospitals are looking into AI solutions to support and perform operational initiatives that increase precision and cost effectiveness. Robotics have been used for disabled and patient care assistance. Medical decision making has been supported through predictive analytics and general healthcare management technology. Network connectivity allows access to affordable healthcare around the globe in a cost-effective way.","PeriodicalId":406435,"journal":{"name":"CompSciRN: Other Machine Learning (Topic)","volume":"122 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128124073","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper we review existing modelling approaches for analysing claims experience in the presence of reporting delays, reviewing the formulation of mortality incidence models such as GLMs. We then show how these approaches have traditionally been adjusted for late reporting of claims using either the IBNR approach or the more recent EBNER approach. We then go on to introduce a new model formulation that combines a model for late reported claims with a model for mortality incidence into a single model formulation. We then illustrate the use and performance of the traditional and the combined model formulations on data from a multinational reinsurer. We show how GLMs, lasso regression, gradient boosted trees and deep learning can be applied to the new formulation to produce results of superior accuracy compared to the traditional approaches.
{"title":"Using Machine Learning to Model Claims Experience and Reporting Delays for Pricing and Reserving","authors":"Louis Rossouw, Ronald Richman","doi":"10.2139/ssrn.3465424","DOIUrl":"https://doi.org/10.2139/ssrn.3465424","url":null,"abstract":"In this paper we review existing modelling approaches for analysing claims experience in the presence of reporting delays, reviewing the formulation of mortality incidence models such as GLMs. We then show how these approaches have traditionally been adjusted for late reporting of claims using either the IBNR approach or the more recent EBNER approach. We then go on to introduce a new model formulation that combines a model for late reported claims with a model for mortality incidence into a single model formulation. We then illustrate the use and performance of the traditional and the combined model formulations on data from a multinational reinsurer. We show how GLMs, lasso regression, gradient boosted trees and deep learning can be applied to the new formulation to produce results of superior accuracy compared to the traditional approaches.","PeriodicalId":406435,"journal":{"name":"CompSciRN: Other Machine Learning (Topic)","volume":"188 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114746049","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}
L. P, Priyanka Prakash, M. M, R. G. Babukarthik, Bonduvenkat B.
Investigated the efficiency of proposed modalities including immunotherapy and cryotherapy for treatment of wart lesions. Cryotherapy with liquid nitrogen is a favorable and different treatment in most patients. A clinical study of efficiency of garlic extract versus cryotherapy in the treatment of male genital wart. With recent technological advancements in data mining and machine learning techniques, early stage of disease can be predicted with a higher degree of accuracy even in the field of medical diagnosis.We proposed Huddle PSO in machine learning using K-means algorithm and Support Vector Machine (SVM). In future we plan apply the proposed work for the treatment of brain tumors.
{"title":"A Novel Approach for Prediction of Warts Disease Treatment Methods: Machine Learning Techniques","authors":"L. P, Priyanka Prakash, M. M, R. G. Babukarthik, Bonduvenkat B.","doi":"10.2139/ssrn.3463673","DOIUrl":"https://doi.org/10.2139/ssrn.3463673","url":null,"abstract":"Investigated the efficiency of proposed modalities including immunotherapy and cryotherapy for treatment of wart lesions. Cryotherapy with liquid nitrogen is a favorable and different treatment in most patients. A clinical study of efficiency of garlic extract versus cryotherapy in the treatment of male genital wart. With recent technological advancements in data mining and machine learning techniques, early stage of disease can be predicted with a higher degree of accuracy even in the field of medical diagnosis.We proposed Huddle PSO in machine learning using K-means algorithm and Support Vector Machine (SVM). In future we plan apply the proposed work for the treatment of brain tumors.","PeriodicalId":406435,"journal":{"name":"CompSciRN: Other Machine Learning (Topic)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122226777","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}
Dr. Maria Papadaki, Dr Nikolaos Bakas, Professor Edward Ochieng, Dr. Ioannis Karamitsos, Dr. Richard Kirkham
The literature review highlights that previous studies have been identifying risk management as an essential tool for project management and could increase the chance of successfully meeting project objectives. In addition, as found from the reviewed literature, risk management has been seen as a tool of allowing the project team to communicate risk information, so as to enhance the decision-making process towards balancing threats and opportunities. Thus, this research aims to examine participants’ views on the alignment of risk management, project management and organizational project success. Machine learning algorithms are employed to explore collective data from posts on twitter in order to obtain valuable knowledge about discussions regarding risk management, and project management. Additionally, the corresponding scientific literature obtained from Scopus database was analyzed utilizing bibliometric tools, in order to investigate diverse perceptions in academia and industry. Findings of this study will have implications for practitioners’ perception of project risk management.
{"title":"Big Data From Social Media and Scientific Literature Databases Reveals Relationships Among Risk Management, Project Management and Project Success","authors":"Dr. Maria Papadaki, Dr Nikolaos Bakas, Professor Edward Ochieng, Dr. Ioannis Karamitsos, Dr. Richard Kirkham","doi":"10.2139/ssrn.3459936","DOIUrl":"https://doi.org/10.2139/ssrn.3459936","url":null,"abstract":"The literature review highlights that previous studies have been identifying risk management as an essential tool for project management and could increase the chance of successfully meeting project objectives. In addition, as found from the reviewed literature, risk management has been seen as a tool of allowing the project team to communicate risk information, so as to enhance the decision-making process towards balancing threats and opportunities. Thus, this research aims to examine participants’ views on the alignment of risk management, project management and organizational project success. Machine learning algorithms are employed to explore collective data from posts on twitter in order to obtain valuable knowledge about discussions regarding risk management, and project management. Additionally, the corresponding scientific literature obtained from Scopus database was analyzed utilizing bibliometric tools, in order to investigate diverse perceptions in academia and industry. Findings of this study will have implications for practitioners’ perception of project risk management.","PeriodicalId":406435,"journal":{"name":"CompSciRN: Other Machine Learning (Topic)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116454659","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}
Abstract We set out in this study to review a vast amount of recent literature on machine learning (ML) approaches to predicting financial distress (FD), including supervised, unsupervised and hybrid supervised–unsupervised learning algorithms. Four supervised ML models including the traditional support vector machine (SVM), recently developed hybrid associative memory with translation (HACT), hybrid GA-fuzzy clustering and extreme gradient boosting (XGBoost) were compared in prediction performance to the unsupervised classifier deep belief network (DBN) and the hybrid DBN-SVM model, whereby a total of sixteen financial variables were selected from the financial statements of the publicly-listed Taiwanese firms as inputs to the six approaches. Our empirical findings, covering the 2010–2016 sample period, demonstrated that among the four supervised algorithms, the XGBoost provided the most accurate FD prediction. Moreover, the hybrid DBN-SVM model was able to generate more accurate forecasts than the use of either the SVM or the classifier DBN in isolation.
{"title":"A New Perspective of Performance Comparison among Machine Learning Algorithms for Financial Distress Prediction","authors":"Yuping Huang, Meng‐Feng Yen","doi":"10.2139/ssrn.3437863","DOIUrl":"https://doi.org/10.2139/ssrn.3437863","url":null,"abstract":"Abstract We set out in this study to review a vast amount of recent literature on machine learning (ML) approaches to predicting financial distress (FD), including supervised, unsupervised and hybrid supervised–unsupervised learning algorithms. Four supervised ML models including the traditional support vector machine (SVM), recently developed hybrid associative memory with translation (HACT), hybrid GA-fuzzy clustering and extreme gradient boosting (XGBoost) were compared in prediction performance to the unsupervised classifier deep belief network (DBN) and the hybrid DBN-SVM model, whereby a total of sixteen financial variables were selected from the financial statements of the publicly-listed Taiwanese firms as inputs to the six approaches. Our empirical findings, covering the 2010–2016 sample period, demonstrated that among the four supervised algorithms, the XGBoost provided the most accurate FD prediction. Moreover, the hybrid DBN-SVM model was able to generate more accurate forecasts than the use of either the SVM or the classifier DBN in isolation.","PeriodicalId":406435,"journal":{"name":"CompSciRN: Other Machine Learning (Topic)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126032255","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The ECG represents a graph of variations in electrical potential generated by the heart and recorded at the body surface. The contraction and relaxation of the cardiac muscles repression in generation of electrical potential which could be used to diagnose different disorders of the heart. In this paper we implement a Novel method to improve the accuracy of heart disease identification using K-NN method. The visual identification of different wavelet features in the ECG signal is done to compare it with respect to the normal values for finding the novelties in it so that physician can make brief move against any issue.
{"title":"A Novel Method to Improve the Accuracy of Heart Disease Identification Using K-NN Method","authors":"Sai Chaitanya, N. Ramaiah","doi":"10.2139/ssrn.3397002","DOIUrl":"https://doi.org/10.2139/ssrn.3397002","url":null,"abstract":"The ECG represents a graph of variations in electrical potential generated by the heart and recorded at the body surface. The contraction and relaxation of the cardiac muscles repression in generation of electrical potential which could be used to diagnose different disorders of the heart. In this paper we implement a Novel method to improve the accuracy of heart disease identification using K-NN method. The visual identification of different wavelet features in the ECG signal is done to compare it with respect to the normal values for finding the novelties in it so that physician can make brief move against any issue.","PeriodicalId":406435,"journal":{"name":"CompSciRN: Other Machine Learning (Topic)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121506690","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}
Savita Choudhary, V. Gaurav, Tushar Sharma, Vishal V, Pradyumna K R
India has witnessed an alarming increase in the number of dengue cases to the count of about 300 percent since 2009 as per the National Health Profile, 2018. Dengue is considered a serious threat not only in India but also is becoming a problem all over the world especially in tropical countries like Indonesia, India and Malaysia. Dengue cases were widespread during the onset and the duration of monsoon due to the collection of water creating breeding grounds for female aedes mosquitoes which are vectors for Flavivirus (Dengue virus). With the lack of appropriate infrastructure and methodology to identify vulnerable regions in India, the cases of dengue have been on the rise. This paper is an attempt to use machine learning and statistical models to predict dengue cases across India and identify the patterns between climatic factors, urbanization and number of cases reported for dengue. This includes the spread spectrum of dengue and also accounts as an AI based mitigative forecast model to alert the concerned authorities before the spread of the epidemic. This will enable the concerned authorities to gauge the situation and take appropriate steps to prevent the pandemy.
{"title":"Forecasting Dengue and Studying its Plausible Pandemy using Machine Learning","authors":"Savita Choudhary, V. Gaurav, Tushar Sharma, Vishal V, Pradyumna K R","doi":"10.2139/ssrn.3507320","DOIUrl":"https://doi.org/10.2139/ssrn.3507320","url":null,"abstract":"India has witnessed an alarming increase in the number of dengue cases to the count of about 300 percent since 2009 as per the National Health Profile, 2018. Dengue is considered a serious threat not only in India but also is becoming a problem all over the world especially in tropical countries like Indonesia, India and Malaysia. Dengue cases were widespread during the onset and the duration of monsoon due to the collection of water creating breeding grounds for female aedes mosquitoes which are vectors for Flavivirus (Dengue virus). With the lack of appropriate infrastructure and methodology to identify vulnerable regions in India, the cases of dengue have been on the rise. This paper is an attempt to use machine learning and statistical models to predict dengue cases across India and identify the patterns between climatic factors, urbanization and number of cases reported for dengue. This includes the spread spectrum of dengue and also accounts as an AI based mitigative forecast model to alert the concerned authorities before the spread of the epidemic. This will enable the concerned authorities to gauge the situation and take appropriate steps to prevent the pandemy.","PeriodicalId":406435,"journal":{"name":"CompSciRN: Other Machine Learning (Topic)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125157302","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}
G. Parthasarathy, L. Ramanathan, Y. Justindhas, J. Saravanakumar, J. Darwin
Today, the total transaction volume of credit cards is increasing consistently, as a result fraudulent transaction cases are also on a rise, producing losses in billions of dollars for financial institutions and banking sectors every year. Hence there is a need for a robust, reliable mechanism which is able to identify and prevent such fraudulent transactions effectively and efficiently. Some data mining techniques helps in detecting patterns between data attributes (classifying the transaction as fraudulent or non-fraudulent) and results in probabilistic prediction of the transaction category. In this study, multiple Machine Learning classification techniques are applied on a highly imbalanced datasets consisting of credit card transaction. ‘Chip and Pin’ is considered as one of the trusted mechanisms today in terms of securing payment transaction but even this mechanism doesn’t stops fake credit card utilizations on virtual Point Of Sale nodes or email orders known as an online 'credit card bankrupt'. It was observed that SVM, Random Forest and J48 Decision Tree classifiers yield a very high accuracy ratio but are suggested not to be leveraged while classifying such dataset where class imbalance is present. While thinking about these methodologies, this investigation gives a comprehensive overview of various classification methods, their highlights and restrictions of bankruptcy.
如今，信用卡交易总量持续增长，因此欺诈交易案件也在增加，每年给金融机构和银行业造成数十亿美元的损失。因此，需要一个强有力的、可靠的机制，能够有效地识别和防止这种欺诈性交易。一些数据挖掘技术有助于检测数据属性之间的模式(将事务分类为欺诈性或非欺诈性)，并对事务类别进行概率预测。在本研究中，将多种机器学习分类技术应用于由信用卡交易组成的高度不平衡数据集。“芯片和密码”被认为是当今安全支付交易的可信机制之一，但即使这种机制也无法阻止虚拟销售点节点或电子邮件订单上的虚假信用卡使用，即在线“信用卡破产”。观察到SVM, Random Forest和J48 Decision Tree分类器产生非常高的准确率，但建议不要在分类存在类不平衡的数据集时使用。在思考这些方法的同时，本调查全面概述了各种分类方法，它们的亮点和破产的限制。
{"title":"Comparative Case Study of Machine Learning Classification Techniques Using Imbalanced Credit Card Fraud Datasets","authors":"G. Parthasarathy, L. Ramanathan, Y. Justindhas, J. Saravanakumar, J. Darwin","doi":"10.2139/ssrn.3351584","DOIUrl":"https://doi.org/10.2139/ssrn.3351584","url":null,"abstract":"Today, the total transaction volume of credit cards is increasing consistently, as a result fraudulent transaction cases are also on a rise, producing losses in billions of dollars for financial institutions and banking sectors every year. Hence there is a need for a robust, reliable mechanism which is able to identify and prevent such fraudulent transactions effectively and efficiently. Some data mining techniques helps in detecting patterns between data attributes (classifying the transaction as fraudulent or non-fraudulent) and results in probabilistic prediction of the transaction category. In this study, multiple Machine Learning classification techniques are applied on a highly imbalanced datasets consisting of credit card transaction. ‘Chip and Pin’ is considered as one of the trusted mechanisms today in terms of securing payment transaction but even this mechanism doesn’t stops fake credit card utilizations on virtual Point Of Sale nodes or email orders known as an online 'credit card bankrupt'. It was observed that SVM, Random Forest and J48 Decision Tree classifiers yield a very high accuracy ratio but are suggested not to be leveraged while classifying such dataset where class imbalance is present. While thinking about these methodologies, this investigation gives a comprehensive overview of various classification methods, their highlights and restrictions of bankruptcy.<br>","PeriodicalId":406435,"journal":{"name":"CompSciRN: Other Machine Learning (Topic)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121123603","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this article, the Universal Approximation Theorem of Artificial Neural Networks (ANNs) is applied to the SABR stochastic volatility model in order to construct highly efficient representations. Initially, the SABR approximation of Hagan et al. [2002] is considered, then a more accurate integration scheme of McGhee [2011] as well as a two factor finite difference scheme. The resulting ANN calculates 10,000 times faster than the finite difference scheme whilst maintaining a high degree of accuracy. As a result, the ANN dispenses with the need for the commonly used SABR Approximation.
{"title":"An Artificial Neural Network Representation of the SABR Stochastic Volatility Model","authors":"W. Mcghee","doi":"10.2139/ssrn.3288882","DOIUrl":"https://doi.org/10.2139/ssrn.3288882","url":null,"abstract":"In this article, the Universal Approximation Theorem of Artificial Neural Networks (ANNs) is applied to the SABR stochastic volatility model in order to construct highly efficient representations. Initially, the SABR approximation of Hagan et al. [2002] is considered, then a more accurate integration scheme of McGhee [2011] as well as a two factor finite difference scheme. The resulting ANN calculates 10,000 times faster than the finite difference scheme whilst maintaining a high degree of accuracy. As a result, the ANN dispenses with the need for the commonly used SABR Approximation.","PeriodicalId":406435,"journal":{"name":"CompSciRN: Other Machine Learning (Topic)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133318485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we present how ideas from artificial intelligence field can be utilized in proxy modeling problem that is faced by actuarial departments of life insurance companies. The current approaches are reviewed, exposing their incapability to fully mimic the complexity and non-linearity of cash-flow projection models. In order to increase the quality of proxy models, we propose to apply selected machine learning algorithms as well as provide an overview of the theory behind them and present the numerical results with a comparison of model errors for different estimators. The study is performed on real data generated by a large reinsurance company. The text can serve as a guideline for companies willing to introduce machine learning algorithms in their proxy modeling processes.
{"title":"Proxy Modeling in Life Insurance Companies With the Use of Machine Learning Algorithms","authors":"Dawid Kopczyk","doi":"10.2139/ssrn.3396481","DOIUrl":"https://doi.org/10.2139/ssrn.3396481","url":null,"abstract":"In this paper, we present how ideas from artificial intelligence field can be utilized in proxy modeling problem that is faced by actuarial departments of life insurance companies. The current approaches are reviewed, exposing their incapability to fully mimic the complexity and non-linearity of cash-flow projection models. In order to increase the quality of proxy models, we propose to apply selected machine learning algorithms as well as provide an overview of the theory behind them and present the numerical results with a comparison of model errors for different estimators. The study is performed on real data generated by a large reinsurance company. The text can serve as a guideline for companies willing to introduce machine learning algorithms in their proxy modeling processes.","PeriodicalId":406435,"journal":{"name":"CompSciRN: Other Machine Learning (Topic)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121208974","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：