Pub Date : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106636
R. Askerov, Eric Kwon, L. Song, Dylan Weber, Oliver Schaer, Faraz Dadgostari, Stephen Adams
Nowadays, financial firms can interpret press releases within few seconds using natural language processing algorithms. Therefore, it is important for public companies to structure its communications in a way that accounts for how the market digests its public information and avoid unnecessary volatility. Companies want to know the impression of their communications, such as investors calls and annual reports, among the investment community including analysts, financial press, and institutional investors. While there have been research papers connecting sentiment analysis of company communication materials to stock movement, research on identifying any similarities in communication styles among public companies has not been a major topic. We aimed to quantify the sentiment of those communication materials and determine if there are any discernible communication styles among leading technology companies. In addition, we conducted analyses and comparisons to stock indices to connect company communication style to market reactions from investors. Our results indicate that there is a signal between sentiment scores derived from Loughran McDonald dictionary and market-residualized stock performance of our company set, highlighting the benefits one can obtain from using NLP techniques.
{"title":"Natural Language Processing for Company Financial Communication Style","authors":"R. Askerov, Eric Kwon, L. Song, Dylan Weber, Oliver Schaer, Faraz Dadgostari, Stephen Adams","doi":"10.1109/SIEDS49339.2020.9106636","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106636","url":null,"abstract":"Nowadays, financial firms can interpret press releases within few seconds using natural language processing algorithms. Therefore, it is important for public companies to structure its communications in a way that accounts for how the market digests its public information and avoid unnecessary volatility. Companies want to know the impression of their communications, such as investors calls and annual reports, among the investment community including analysts, financial press, and institutional investors. While there have been research papers connecting sentiment analysis of company communication materials to stock movement, research on identifying any similarities in communication styles among public companies has not been a major topic. We aimed to quantify the sentiment of those communication materials and determine if there are any discernible communication styles among leading technology companies. In addition, we conducted analyses and comparisons to stock indices to connect company communication style to market reactions from investors. Our results indicate that there is a signal between sentiment scores derived from Loughran McDonald dictionary and market-residualized stock performance of our company set, highlighting the benefits one can obtain from using NLP techniques.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115337291","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106640
Camryn Burley, Darby Anderson, Amanda Brownlee, Georgie Lafer, Taylor Luong, Meaghan McGowan, Judy Nguyen, William Trotter, Halle Wine, Anna N. Baglione, Laura E. Barnes
Approximately one in five adults in the United States have been diagnosed with some form of mental illness, but less than half received treatment in this past year [1]. An interdisciplinary team at the University of Virginia aims to reduce this gap in mental health coverage through its freely accessible online research platform, the MindTrails Project. The MindTrails Calm Thinking study evaluates cognitive bias modification for interpretation (CBM-I), an intervention that aims to reframe the thinking patterns of highly anxious individuals when they respond to ambiguous situations that they might interpret as stressful. MindTrails is experiencing a high attrition (dropout) rate, which is common to eHealth interventions. In response to this, our project utilized two novel approaches to online anxiety interventions to improve engagement and retention: (1) personalization of training content and (2) implementation intentions and goal setting. We designed a prototype for a new mobile interface that engages users with a journal to record implementation intentions and goals. Users also have the ability to choose the domain of anxiety (e.g., relationships, health) that they would like to work on. To further incorporate these psychological principles into the MindTrails program, suggestions for future work are also discussed. We hypothesize that, with its new user-centered mobile interface, the Calm Thinking mobile application will further connect users with an evidence-based mental health intervention and increase the efficacy of the program.
在美国,大约五分之一的成年人被诊断出患有某种形式的精神疾病,但在过去的一年中,接受治疗的不到一半[1]。弗吉尼亚大学(University of Virginia)的一个跨学科团队旨在通过其免费访问的在线研究平台MindTrails Project,缩小心理健康覆盖方面的差距。MindTrails Calm Thinking研究评估了认知偏差修正解释(CBM-I),这是一种干预措施,旨在重新构建高度焦虑的个体在他们可能将其解释为压力的模糊情况下的思维模式。MindTrails正在经历高流失率,这在电子健康干预中很常见。针对这一点,我们的项目采用了两种新颖的在线焦虑干预方法来提高参与度和留存率:(1)个性化培训内容;(2)实施意图和目标设定。我们为新的移动界面设计了一个原型,让用户通过日志记录执行意图和目标。用户还可以选择他们想要处理的焦虑领域(例如,人际关系、健康)。为了进一步将这些心理学原理纳入MindTrails计划,还讨论了对未来工作的建议。我们假设,通过其新的以用户为中心的移动界面,Calm Thinking移动应用程序将进一步将用户与基于证据的心理健康干预联系起来,并提高程序的有效性。
{"title":"Increasing Engagement in eHealth Interventions Using Personalization and Implementation Intentions","authors":"Camryn Burley, Darby Anderson, Amanda Brownlee, Georgie Lafer, Taylor Luong, Meaghan McGowan, Judy Nguyen, William Trotter, Halle Wine, Anna N. Baglione, Laura E. Barnes","doi":"10.1109/SIEDS49339.2020.9106640","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106640","url":null,"abstract":"Approximately one in five adults in the United States have been diagnosed with some form of mental illness, but less than half received treatment in this past year [1]. An interdisciplinary team at the University of Virginia aims to reduce this gap in mental health coverage through its freely accessible online research platform, the MindTrails Project. The MindTrails Calm Thinking study evaluates cognitive bias modification for interpretation (CBM-I), an intervention that aims to reframe the thinking patterns of highly anxious individuals when they respond to ambiguous situations that they might interpret as stressful. MindTrails is experiencing a high attrition (dropout) rate, which is common to eHealth interventions. In response to this, our project utilized two novel approaches to online anxiety interventions to improve engagement and retention: (1) personalization of training content and (2) implementation intentions and goal setting. We designed a prototype for a new mobile interface that engages users with a journal to record implementation intentions and goals. Users also have the ability to choose the domain of anxiety (e.g., relationships, health) that they would like to work on. To further incorporate these psychological principles into the MindTrails program, suggestions for future work are also discussed. We hypothesize that, with its new user-centered mobile interface, the Calm Thinking mobile application will further connect users with an evidence-based mental health intervention and increase the efficacy of the program.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115277668","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106585
Caroline G. George, Declan R. Tyranski, Devin P. Simons, Jameson D. O’Quinn, Emily York, A. Salman
Smart homes are becoming increasingly common across the United States with the advent and ongoing development of new and improved IoT devices. With the increasing use of these devices, there has been exponential growth in the amount of data collected on individuals. This poses potential privacy risks that can affect the lives of these users and raises an ethical question about what information is being collected and how it is used. Much of the data collected is used to discern information about when, why, and how the device was used. It must be questioned how much information collection is necessary and at what point does it become a violation of privacy. Additionally, many devices collect data at various times when the user may not be aware of it. This information collected can be used to target specific advertisements, influence users, or be sold to third-party sources. Although much of this information is laid out in lengthy, often deceptive terms and conditions for most devices, many people do not read them or understand the implications they pose. In this paper, we present a solution that monitors data leaving the house through a device integrated within the home network with the aim of spreading awareness surrounding the potential risks associated with this issue and to work towards limiting the amount of information that is collected.
{"title":"Integrating Social and Technical Solutions to Address Privacy in Smart Homes","authors":"Caroline G. George, Declan R. Tyranski, Devin P. Simons, Jameson D. O’Quinn, Emily York, A. Salman","doi":"10.1109/SIEDS49339.2020.9106585","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106585","url":null,"abstract":"Smart homes are becoming increasingly common across the United States with the advent and ongoing development of new and improved IoT devices. With the increasing use of these devices, there has been exponential growth in the amount of data collected on individuals. This poses potential privacy risks that can affect the lives of these users and raises an ethical question about what information is being collected and how it is used. Much of the data collected is used to discern information about when, why, and how the device was used. It must be questioned how much information collection is necessary and at what point does it become a violation of privacy. Additionally, many devices collect data at various times when the user may not be aware of it. This information collected can be used to target specific advertisements, influence users, or be sold to third-party sources. Although much of this information is laid out in lengthy, often deceptive terms and conditions for most devices, many people do not read them or understand the implications they pose. In this paper, we present a solution that monitors data leaving the house through a device integrated within the home network with the aim of spreading awareness surrounding the potential risks associated with this issue and to work towards limiting the amount of information that is collected.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115881298","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106628
Melissa C Phillips, Rebecca Stein, Taeheon Park
Image classification and object recognition with neural networks could have applications in aesthetically-focused branches of the humanities, such as landscape architecture. However, such methods require either the assembly of a massive, domain specific labeled data set or use of network weights initialized on another data set, a technique known as transfer learning. Transfer learning research has established that a pre-trained convolutional neural network (CNN) can achieve high accuracy on new image recognition tasks with relatively few training images. In practice, pre-trained tends to mean pre-trained on ImageNet, the standard dataset for computer vision research. Experiments have shown that the dataset on which a pre-trained model was originally optimized can quantitatively bias it. The goal of this project was to design an experiment to qualitatively analyze how the dataset used to initialize a pre-trained classification system affects its behavior at progressive network layers using feature visualization strategies. We initialized two ResNet-18 CNNs with weights pre-trained on ImageNet and the Places365 dataset, respectively, and fine-tuned them for a new classification task on a landscape image dataset which we collected. Using class activation optimization methods taken from the deep visualization literature, we compared the network filters at several hidden layers and the final output layers. The class activation optimization results show that even at early stages in the networks, their neurons exhibit notably different behavior. Accordingly, we show both that feature visualization techniques can be used to qualitatively study the effect of original training data on transfer learning and, consequently, that the homogeneous use of ImageNet in computer vision experiments may have notable implications for model behavior.
{"title":"Analyzing Pre-Trained Neural Network Behavior with Layer Activation Optimization","authors":"Melissa C Phillips, Rebecca Stein, Taeheon Park","doi":"10.1109/SIEDS49339.2020.9106628","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106628","url":null,"abstract":"Image classification and object recognition with neural networks could have applications in aesthetically-focused branches of the humanities, such as landscape architecture. However, such methods require either the assembly of a massive, domain specific labeled data set or use of network weights initialized on another data set, a technique known as transfer learning. Transfer learning research has established that a pre-trained convolutional neural network (CNN) can achieve high accuracy on new image recognition tasks with relatively few training images. In practice, pre-trained tends to mean pre-trained on ImageNet, the standard dataset for computer vision research. Experiments have shown that the dataset on which a pre-trained model was originally optimized can quantitatively bias it. The goal of this project was to design an experiment to qualitatively analyze how the dataset used to initialize a pre-trained classification system affects its behavior at progressive network layers using feature visualization strategies. We initialized two ResNet-18 CNNs with weights pre-trained on ImageNet and the Places365 dataset, respectively, and fine-tuned them for a new classification task on a landscape image dataset which we collected. Using class activation optimization methods taken from the deep visualization literature, we compared the network filters at several hidden layers and the final output layers. The class activation optimization results show that even at early stages in the networks, their neurons exhibit notably different behavior. Accordingly, we show both that feature visualization techniques can be used to qualitatively study the effect of original training data on transfer learning and, consequently, that the homogeneous use of ImageNet in computer vision experiments may have notable implications for model behavior.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132615672","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106653
W. Donohue, Z. Afridi, K. Sokolyuk, Tyler Bedwell, Emily York, A. Salman
A cashless society is an economic state which handles financial transactions not in the form of traditional mediums of currency, such as cash or coins, but by transferring digital data (usually by electronic means, such as credit cards and mobile data) between participating parties.Participants of a cashless society must Figure out a way to protect their transaction data, acknowledging the risks of organizations collecting mass amounts of said data, which result in a reduction of personal privacy. Balancing individual privacy with data security is vital in the information age, especially considering the increasing risk of data breaches and exploitation.In order to increase privacy in a cashless society, a few courses of action can be combined to produce a lasting and desirable result for users: A new kind of banking service that assigns randomized numbers to credit cards, the use of blockchain to monitor all transactions from individuals, and a campaign to educate and inform key stakeholders about security and privacy risks to provide the necessary tools and background knowledge to safeguard their own information before interaction with a foreign entity or other third parties (i.e. cybersecurity departments, IT technicians, etc). Blockchain and card number randomization are both susceptible to zero-day errors, bugs, and varied levels of social acceptance. This preliminary research draws on a systems analysis of cashless systems to identify and analyze a set of social and technical solutions to support a robust cashless system that protects users’ privacy and maintains the security of the system.The information found and analyzed will be beneficial by exposing weak points in current methods of data integrity and security. Learning about current and future methods of managing privacy and data security in the technological age would be helpful in creating preventative countermeasures. This study provides critical steps to prevent the loss of personal privacy in a cashless system.
{"title":"Cashless Society: Managing Privacy and Security in the Technological Age","authors":"W. Donohue, Z. Afridi, K. Sokolyuk, Tyler Bedwell, Emily York, A. Salman","doi":"10.1109/SIEDS49339.2020.9106653","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106653","url":null,"abstract":"A cashless society is an economic state which handles financial transactions not in the form of traditional mediums of currency, such as cash or coins, but by transferring digital data (usually by electronic means, such as credit cards and mobile data) between participating parties.Participants of a cashless society must Figure out a way to protect their transaction data, acknowledging the risks of organizations collecting mass amounts of said data, which result in a reduction of personal privacy. Balancing individual privacy with data security is vital in the information age, especially considering the increasing risk of data breaches and exploitation.In order to increase privacy in a cashless society, a few courses of action can be combined to produce a lasting and desirable result for users: A new kind of banking service that assigns randomized numbers to credit cards, the use of blockchain to monitor all transactions from individuals, and a campaign to educate and inform key stakeholders about security and privacy risks to provide the necessary tools and background knowledge to safeguard their own information before interaction with a foreign entity or other third parties (i.e. cybersecurity departments, IT technicians, etc). Blockchain and card number randomization are both susceptible to zero-day errors, bugs, and varied levels of social acceptance. This preliminary research draws on a systems analysis of cashless systems to identify and analyze a set of social and technical solutions to support a robust cashless system that protects users’ privacy and maintains the security of the system.The information found and analyzed will be beneficial by exposing weak points in current methods of data integrity and security. Learning about current and future methods of managing privacy and data security in the technological age would be helpful in creating preventative countermeasures. This study provides critical steps to prevent the loss of personal privacy in a cashless system.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124925368","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106665
Aria Amini, Hamza Abshir, Kamilla Quinones Burgos, Mahmoud Moharrem, Sara Elkholy
Hiring Managers at System Engineering companies have to select engineering candidates that will be value-added to the organization now and in the future and avoid costly bad hires. Interviews with Hiring Managers identified that they use Grade Point Average (GPA), work experience (e.g. internships), and skills (e.g. programming languages) to choose candidates for interview. To reduce their risk, they also use Professional Licenses as a discriminator. For entry-level Systems Engineers, the Associate Systems Engineering Professional (ASEP) certificate offered by INCOSE is the appropriate Professional License. A passing grade is 70% and only 60% of the people taking the exam pass. A tutorial system for students taking the exam is needed to minimize the risk of not passing the exam (i.e. guarantee passing the exam), reducing the time to study for the exam, and to make studying for the exam an enjoyable experience. The Concept-of-Operations for the Tutorial System is to assess student's knowledge with a diagnostic quiz, provide practice quizzes and supplemental materials, and evaluate students' performance with assessment quizzes. The tutorial is self-paced and includes repetition to avoid forgetting. The Tutorial System was implemented in Google Classroom In addition to Google classroom, an application called OwlCamp was created to provide the practice quizzes for the students. The Google Classroom learning management and OwlCamp have undergone verification testing and both satisfy the mission and design requirements. A Validation Test of the Tutorial System was conducted. Seventeen Senior Systems Engineering students were given a Diagnostic Test each week followed by supplemental learning materials, ending with assessment quizzes to test their knowledge. The null hypothesis tested is: “The ASEP Tutorial System will not improve the students’ grade between the diagnostic test and the Assessment quiz Using a 5% level of significance, the data shows that there is indeed a difference between diagnostics and assessments. A 5yearprojection with 10% market penetration for annual market size of 2250 SE students per year, generates cumulative revenue of $675,000. With nonrecurring development and testing cost of $75,205, and recurring maintenance costs of $1281 per year, the 5 year profit is estimated at $3,293,390. The 5 year ROI is 112.95% and the Break-even is in year 1.
{"title":"Design of a Tutorial System for the Associate Systems Engineering Professional (ASEP) Exam","authors":"Aria Amini, Hamza Abshir, Kamilla Quinones Burgos, Mahmoud Moharrem, Sara Elkholy","doi":"10.1109/SIEDS49339.2020.9106665","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106665","url":null,"abstract":"Hiring Managers at System Engineering companies have to select engineering candidates that will be value-added to the organization now and in the future and avoid costly bad hires. Interviews with Hiring Managers identified that they use Grade Point Average (GPA), work experience (e.g. internships), and skills (e.g. programming languages) to choose candidates for interview. To reduce their risk, they also use Professional Licenses as a discriminator. For entry-level Systems Engineers, the Associate Systems Engineering Professional (ASEP) certificate offered by INCOSE is the appropriate Professional License. A passing grade is 70% and only 60% of the people taking the exam pass. A tutorial system for students taking the exam is needed to minimize the risk of not passing the exam (i.e. guarantee passing the exam), reducing the time to study for the exam, and to make studying for the exam an enjoyable experience. The Concept-of-Operations for the Tutorial System is to assess student's knowledge with a diagnostic quiz, provide practice quizzes and supplemental materials, and evaluate students' performance with assessment quizzes. The tutorial is self-paced and includes repetition to avoid forgetting. The Tutorial System was implemented in Google Classroom In addition to Google classroom, an application called OwlCamp was created to provide the practice quizzes for the students. The Google Classroom learning management and OwlCamp have undergone verification testing and both satisfy the mission and design requirements. A Validation Test of the Tutorial System was conducted. Seventeen Senior Systems Engineering students were given a Diagnostic Test each week followed by supplemental learning materials, ending with assessment quizzes to test their knowledge. The null hypothesis tested is: “The ASEP Tutorial System will not improve the students’ grade between the diagnostic test and the Assessment quiz Using a 5% level of significance, the data shows that there is indeed a difference between diagnostics and assessments. A 5yearprojection with 10% market penetration for annual market size of 2250 SE students per year, generates cumulative revenue of $675,000. With nonrecurring development and testing cost of $75,205, and recurring maintenance costs of $1281 per year, the 5 year profit is estimated at $3,293,390. The 5 year ROI is 112.95% and the Break-even is in year 1.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125540100","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106586
Mritika Contractor, Gabriella Luna, Shreya N. Patel, Sophie Steinberg
There is an increasing trend among urban populations to recognize the importance of fresh produce in their diets and its impact on reducing the carbon footprint created by food transportation. Thus, urban farming as a produce source has grown in popularity in recent years. One farming method that is gaining attention is urban rooftop farming, which integrates farming practices into city infrastructure without requiring expensive real estate or large warehouse-type structures with interior grow lighting. Rooftops in American cities represent the largest unoccupied urban space for agricultural purposes, but they remain underutilized. Selection of feasible and safe locations, obtaining permissions, designing and constructing the farm itself, selecting appropriate crops, and projecting farm outputs are all complex issues that impede the adoption of rooftop farming. To address such complexity, this project developed a prototype Decision Support and Planning Tool that assesses rooftop feasibility, supports informed and geographically appropriate rooftop farm design and crop selection, and predicts crop yield. The team implemented requirements analysis and functional decomposition to identify structural, safety and access requirements for rooftop farming. A second phase of the requirements analysis and functional decomposition was performed to identify agricultural methods and farm design. As a result, “square foot farming” was selected as the appropriate basis for farm and tool design. Users are also guided to input their desired level of effort for maintenance, time to maturity, and crop yield to identify crops most suitable to the specific rooftop location. Analytic hierarchy process (AHP) was used to scale and calculate the weights associated with the users’ maintenance preferences. A linear programming model based on knapsack optimization was used to project maximum total yield based on available square footage and crop yield preferences. Two proof-of-concept rooftop farms, generated by the prototype Decision Support and Planning Tool, were constructed in Washington, DC and Los Angeles. Prior to the spread of COVID-19, these farms were intended to validate model results against actual yield from crops produced over a 90day growing horizon. Instead, the farms validated rooftop assessment and crop selection tool functions.
{"title":"Decision Support and Planning Tool to Facilitate Urban Rooftop Farming","authors":"Mritika Contractor, Gabriella Luna, Shreya N. Patel, Sophie Steinberg","doi":"10.1109/SIEDS49339.2020.9106586","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106586","url":null,"abstract":"There is an increasing trend among urban populations to recognize the importance of fresh produce in their diets and its impact on reducing the carbon footprint created by food transportation. Thus, urban farming as a produce source has grown in popularity in recent years. One farming method that is gaining attention is urban rooftop farming, which integrates farming practices into city infrastructure without requiring expensive real estate or large warehouse-type structures with interior grow lighting. Rooftops in American cities represent the largest unoccupied urban space for agricultural purposes, but they remain underutilized. Selection of feasible and safe locations, obtaining permissions, designing and constructing the farm itself, selecting appropriate crops, and projecting farm outputs are all complex issues that impede the adoption of rooftop farming. To address such complexity, this project developed a prototype Decision Support and Planning Tool that assesses rooftop feasibility, supports informed and geographically appropriate rooftop farm design and crop selection, and predicts crop yield. The team implemented requirements analysis and functional decomposition to identify structural, safety and access requirements for rooftop farming. A second phase of the requirements analysis and functional decomposition was performed to identify agricultural methods and farm design. As a result, “square foot farming” was selected as the appropriate basis for farm and tool design. Users are also guided to input their desired level of effort for maintenance, time to maturity, and crop yield to identify crops most suitable to the specific rooftop location. Analytic hierarchy process (AHP) was used to scale and calculate the weights associated with the users’ maintenance preferences. A linear programming model based on knapsack optimization was used to project maximum total yield based on available square footage and crop yield preferences. Two proof-of-concept rooftop farms, generated by the prototype Decision Support and Planning Tool, were constructed in Washington, DC and Los Angeles. Prior to the spread of COVID-19, these farms were intended to validate model results against actual yield from crops produced over a 90day growing horizon. Instead, the farms validated rooftop assessment and crop selection tool functions.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127100878","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106669
Cedric Harper, Brigitte Hogan, Briana K. Wright
Can private location data be used for the public good? During an emergency, cities and municipalities must disperse limited resources to the areas of greatest need. The data which can best inform these decisions may be hidden within the mobile apps that city residents use on an everyday basis. Given the ethical concerns surrounding location tracking, we address this question using data from X-Mode Social, Inc., a start-up company with open and transparent data sharing policies. X-Mode’s high-quality location data are compliant with both regulations in the European Union (GDPR) and the United States (CCPA). We narrowed our focus to the City of Jacksonville, Florida, which issued mandatory evacuations prior to Hurricane Dorian’s approach in early September 2019. After validating that X-Mode’s data correlates with local population densities, we visualized locations pre- and post-hurricane in order to establish whether mobile app users were able to heed government warnings. Next, we used a combination of both spatial analysis and generalized linear modeling methods to characterize patterns of movement during the evacuation. Finally, we built an interactive web-based app to reveal areas where the evacuation process could potentially be improved. Our results work to fill current knowledge gaps and provide a process with which city and municipal managers might utilize to more effectively allocate resources during a crisis.
{"title":"Applying Mobile Location Data to Improve Hurricane Evacuation Plans","authors":"Cedric Harper, Brigitte Hogan, Briana K. Wright","doi":"10.1109/SIEDS49339.2020.9106669","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106669","url":null,"abstract":"Can private location data be used for the public good? During an emergency, cities and municipalities must disperse limited resources to the areas of greatest need. The data which can best inform these decisions may be hidden within the mobile apps that city residents use on an everyday basis. Given the ethical concerns surrounding location tracking, we address this question using data from X-Mode Social, Inc., a start-up company with open and transparent data sharing policies. X-Mode’s high-quality location data are compliant with both regulations in the European Union (GDPR) and the United States (CCPA). We narrowed our focus to the City of Jacksonville, Florida, which issued mandatory evacuations prior to Hurricane Dorian’s approach in early September 2019. After validating that X-Mode’s data correlates with local population densities, we visualized locations pre- and post-hurricane in order to establish whether mobile app users were able to heed government warnings. Next, we used a combination of both spatial analysis and generalized linear modeling methods to characterize patterns of movement during the evacuation. Finally, we built an interactive web-based app to reveal areas where the evacuation process could potentially be improved. Our results work to fill current knowledge gaps and provide a process with which city and municipal managers might utilize to more effectively allocate resources during a crisis.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128439652","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106583
P. Finley, Grayson Gatti, J. Goodall, Mac Nelson, Kiri Nicholson, K. Shah
Flooding events are expected to increase due to climate change. Because of this, cities across the country need to implement flood mitigation strategies in order to ensure the safety and health of their residents. These cities need improved modeling and sensing capabilities to determine which areas (streets, residential neighborhoods, etc.) are flooding in real-time or are vulnerable to flooding from extreme weather events. Both an objective way to monitor stormwater structures and a methodology to rank such structures in accordance to maintenance needs would be valuable. To rank storm structures by peak flow, the methodology consists of using geographic information system (GIS) data combined with Arc Hydro tools to calculate the peak flow of inlet structures grouped by diameter via the rational method. The sensing system is an optical sensor that communicates using LoRa to a The Things Network node. A virtual machine running a Python script extracts the data from The Things Network and places it in an SQLite3 database that can be used for visualization and analysis by decision-makers. Both the GIS-based stormwater infrastructure assessment methodology and flood sensor system are demonstrated using neighborhoods in the City of Charlottesville as a case study.
由于气候变化,洪水事件预计会增加。因此,全国各城市需要实施防洪战略,以确保居民的安全和健康。这些城市需要改进建模和传感能力,以确定哪些区域(街道、居民区等)正在实时发生洪水,或容易受到极端天气事件的洪水影响。一种客观的方法来监测雨水结构,以及一种根据维修需要对这些结构进行排序的方法,都是很有价值的。采用地理信息系统(GIS)数据与Arc Hydro工具相结合的方法,通过合理的方法计算按直径分组的入口结构的峰值流量。传感系统是一个光学传感器,通过LoRa与The Things Network节点通信。运行Python脚本的虚拟机从the Things Network中提取数据并将其放入SQLite3数据库中,该数据库可用于决策者的可视化和分析。以夏洛茨维尔市的社区为例,展示了基于gis的雨水基础设施评估方法和洪水传感器系统。
{"title":"Flood Monitoring and Mitigation Strategies for Flood-Prone Urban Areas","authors":"P. Finley, Grayson Gatti, J. Goodall, Mac Nelson, Kiri Nicholson, K. Shah","doi":"10.1109/SIEDS49339.2020.9106583","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106583","url":null,"abstract":"Flooding events are expected to increase due to climate change. Because of this, cities across the country need to implement flood mitigation strategies in order to ensure the safety and health of their residents. These cities need improved modeling and sensing capabilities to determine which areas (streets, residential neighborhoods, etc.) are flooding in real-time or are vulnerable to flooding from extreme weather events. Both an objective way to monitor stormwater structures and a methodology to rank such structures in accordance to maintenance needs would be valuable. To rank storm structures by peak flow, the methodology consists of using geographic information system (GIS) data combined with Arc Hydro tools to calculate the peak flow of inlet structures grouped by diameter via the rational method. The sensing system is an optical sensor that communicates using LoRa to a The Things Network node. A virtual machine running a Python script extracts the data from The Things Network and places it in an SQLite3 database that can be used for visualization and analysis by decision-makers. Both the GIS-based stormwater infrastructure assessment methodology and flood sensor system are demonstrated using neighborhoods in the City of Charlottesville as a case study.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131180063","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106691
Yunxuan He, Ying Xiong, Y. Tsai
Customer churn prediction plays an important role in business success for insurance companies like Markel Corporation. Each year Markel loses premium because some of their customers choose not to renew their policies. Based on the fact that the cost of attracting new customers is much greater than that of retaining existing customers, it is important for Markel to take early action to engage their customers before a policy expires. The goal in this work is to apply various machine learning methods and obtain an optimal model to predict customer churn rate. The dataset includes customer demographics features, customer behavior features, and macro environmental features. Exploratory analysis is conducted on critical features including policy length and types of coverage to draw insight about the impact of these features on the target variable – customers renew or do not renew their policies. With a large dataset, one of the main challenges is conducting feature dimension reduction and extracting important features to be used with a set of potential ML models. It turns out that the ML model with the best performance on the Area Under the Curve (AUC) metric is Extremely Randomized Trees Classifier and Gradient Boosting Model. Some suggestions on additional features to be incorporated are provided in the final comments. These features will improve predictive performance for the ML model of customer churn for Markel Corporation.
{"title":"Machine Learning Based Approaches to Predict Customer Churn for an Insurance Company","authors":"Yunxuan He, Ying Xiong, Y. Tsai","doi":"10.1109/SIEDS49339.2020.9106691","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106691","url":null,"abstract":"Customer churn prediction plays an important role in business success for insurance companies like Markel Corporation. Each year Markel loses premium because some of their customers choose not to renew their policies. Based on the fact that the cost of attracting new customers is much greater than that of retaining existing customers, it is important for Markel to take early action to engage their customers before a policy expires. The goal in this work is to apply various machine learning methods and obtain an optimal model to predict customer churn rate. The dataset includes customer demographics features, customer behavior features, and macro environmental features. Exploratory analysis is conducted on critical features including policy length and types of coverage to draw insight about the impact of these features on the target variable – customers renew or do not renew their policies. With a large dataset, one of the main challenges is conducting feature dimension reduction and extracting important features to be used with a set of potential ML models. It turns out that the ML model with the best performance on the Area Under the Curve (AUC) metric is Extremely Randomized Trees Classifier and Gradient Boosting Model. Some suggestions on additional features to be incorporated are provided in the final comments. These features will improve predictive performance for the ML model of customer churn for Markel Corporation.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115437334","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}