C. Soares, Philicity Williams, J. Gilbert, G. Dozier
Due to substantial increases in data acquisition and storage, data pre-processing techniques such as feature selection have become increasingly popular in classification tasks. This research proposes a new feature selection algorithm, Class-specific Ensemble Feature Selection (CEFS), which finds class-specific subsets of features optimal to each available classification in the dataset. Each subset is then combined with a classifier to create an ensemble feature selection model which is further used to predict unseen instances. CEFS attempts to provide the diversity and base classifier disagreement sought after in effective ensemble models by providing highly useful, yet highly exclusive feature subsets. Also, the use of a wrapper method gives each subset the chance to perform optimally under the respective base classifier. Preliminary experiments implementing this innovative approach suggest potential improvements of more than 10% over existing methods.
{"title":"A class-specific ensemble feature selection approach for classification problems","authors":"C. Soares, Philicity Williams, J. Gilbert, G. Dozier","doi":"10.1145/1900008.1900054","DOIUrl":"https://doi.org/10.1145/1900008.1900054","url":null,"abstract":"Due to substantial increases in data acquisition and storage, data pre-processing techniques such as feature selection have become increasingly popular in classification tasks. This research proposes a new feature selection algorithm, Class-specific Ensemble Feature Selection (CEFS), which finds class-specific subsets of features optimal to each available classification in the dataset. Each subset is then combined with a classifier to create an ensemble feature selection model which is further used to predict unseen instances. CEFS attempts to provide the diversity and base classifier disagreement sought after in effective ensemble models by providing highly useful, yet highly exclusive feature subsets. Also, the use of a wrapper method gives each subset the chance to perform optimally under the respective base classifier. Preliminary experiments implementing this innovative approach suggest potential improvements of more than 10% over existing methods.","PeriodicalId":333104,"journal":{"name":"ACM SE '10","volume":"110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127621287","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 science and engineering, modeling and simulations are popularly used to gather knowledge and explain a complicated phenomena. These models are typically represented in partial differential equations (PDEs) which can be solved using meshes and sparse matrices. The solution cost of PDEs are dominated by the solution cost of sparse linear systems. Therefore, an efficient sparse linear solver becomes more important with the popularity of scientific modeling and simulations. In this paper, we proposed a robust incomplete LU preconditioning algorithm using constraints diagonal Markowitz. Experimental results with various linear systems show that ILU algorithm using constraints diagonal Markowitz shows better performance and more stable than traditional ILU algorithm with predefined ordering.
{"title":"A robust ILU preconditioner using constraints diagonal Markowitz","authors":"Ingyu Lee","doi":"10.1145/1900008.1900031","DOIUrl":"https://doi.org/10.1145/1900008.1900031","url":null,"abstract":"In science and engineering, modeling and simulations are popularly used to gather knowledge and explain a complicated phenomena. These models are typically represented in partial differential equations (PDEs) which can be solved using meshes and sparse matrices. The solution cost of PDEs are dominated by the solution cost of sparse linear systems. Therefore, an efficient sparse linear solver becomes more important with the popularity of scientific modeling and simulations. In this paper, we proposed a robust incomplete LU preconditioning algorithm using constraints diagonal Markowitz. Experimental results with various linear systems show that ILU algorithm using constraints diagonal Markowitz shows better performance and more stable than traditional ILU algorithm with predefined ordering.","PeriodicalId":333104,"journal":{"name":"ACM SE '10","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123509440","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}
This paper describes a novel tool for analysis of polygonal surface meshes, based on a modified breadth-first traversal of mesh vertices. The value of this technique lies in the fact that it relies only on mesh connectivity, making it invariant to model pose and transformation. It does this by leveraging the close relationship between geometry and connectivity in polygonal meshes. A breadth-first traversal of the mesh is used to construct a simplified graph representation of the mesh, where each vertex in the new graph is associated with some subset of vertices in the original mesh. This graph provides valuable information about the nature of the mesh's shape. This paper discusses the modified breadth-first traversal, the construction of the new graph and some ways in which this graph can be used for shape analysis on the original mesh.
{"title":"Mesh analysis via breadth-first traversal","authors":"David O'Gwynn, J. Johnstone","doi":"10.1145/1900008.1900102","DOIUrl":"https://doi.org/10.1145/1900008.1900102","url":null,"abstract":"This paper describes a novel tool for analysis of polygonal surface meshes, based on a modified breadth-first traversal of mesh vertices. The value of this technique lies in the fact that it relies only on mesh connectivity, making it invariant to model pose and transformation. It does this by leveraging the close relationship between geometry and connectivity in polygonal meshes. A breadth-first traversal of the mesh is used to construct a simplified graph representation of the mesh, where each vertex in the new graph is associated with some subset of vertices in the original mesh. This graph provides valuable information about the nature of the mesh's shape. This paper discusses the modified breadth-first traversal, the construction of the new graph and some ways in which this graph can be used for shape analysis on the original mesh.","PeriodicalId":333104,"journal":{"name":"ACM SE '10","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125225068","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}
Chunking algorithms play an important role in hash-based data de-duplication systems. The Basic Sliding Window (BSW) algorithm is the first prototype of a content-based chunking algorithm that can handle most types of data. The Two Thresholds Two Divisors (TTTD) algorithm was proposed to improve the BSW algorithm by controlling the chunk-size variations. We conducted a series of systematic experiments to evaluate the performances of these two algorithms. We also proposed a new improvement for the TTTD algorithm. Our new approach reduced about 6% of the running time and 50% of the large-sized chunks, and also brought other significant benefits.
{"title":"A running time improvement for the two thresholds two divisors algorithm","authors":"Teng-Sheng Moh, BingChun Chang","doi":"10.1145/1900008.1900101","DOIUrl":"https://doi.org/10.1145/1900008.1900101","url":null,"abstract":"Chunking algorithms play an important role in hash-based data de-duplication systems. The Basic Sliding Window (BSW) algorithm is the first prototype of a content-based chunking algorithm that can handle most types of data. The Two Thresholds Two Divisors (TTTD) algorithm was proposed to improve the BSW algorithm by controlling the chunk-size variations. We conducted a series of systematic experiments to evaluate the performances of these two algorithms. We also proposed a new improvement for the TTTD algorithm. Our new approach reduced about 6% of the running time and 50% of the large-sized chunks, and also brought other significant benefits.","PeriodicalId":333104,"journal":{"name":"ACM SE '10","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129720165","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}
This paper describes our experience of using open source software systems in teaching a graduate level software engineering course. The motivation of this course, the course structure, the assessment, and the outcomes are discussed. The comparative results using different approaches are also presented.
{"title":"Teaching software engineering using open source software","authors":"Guangming Xing","doi":"10.1145/1900008.1900085","DOIUrl":"https://doi.org/10.1145/1900008.1900085","url":null,"abstract":"This paper describes our experience of using open source software systems in teaching a graduate level software engineering course. The motivation of this course, the course structure, the assessment, and the outcomes are discussed. The comparative results using different approaches are also presented.","PeriodicalId":333104,"journal":{"name":"ACM SE '10","volume":"54 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114011321","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 need for power conservation in today's world is continually growing in importance due to increasing energy costs and the desire to conserve natural resources. Energy conservation extends into computer systems because of these systems' performance demands, energy costs, and the growth of environmental awareness. Meanwhile, computer virtualization is growing in popularity in today's computer systems because it allows for the consolidation of multiple servers onto one larger server. Data centers and computational clusters use virtualization because of the many benefits it offers over the use of traditional stand-alone servers such as ease of management, enhanced security, and reduced costs. Virtual machines can be consolidated to obtain maximum power efficiency by taking into account the attributes of applications running in virtual machines. The contribution of this paper is the study of the relationship between applications running in virtual machines, and the effect they have on power consumption. Optimal power efficiency is obtained when virtual machines are consolidated so that all the resources for a given host are fully utilized. This will be used in the goal of building an energy efficient system model and load-balancing algorithm for use in data centers and computational clusters.
{"title":"Towards power efficient consolidation and distribution of virtual machines","authors":"C. Humphries, P. Ruth","doi":"10.1145/1900008.1900109","DOIUrl":"https://doi.org/10.1145/1900008.1900109","url":null,"abstract":"The need for power conservation in today's world is continually growing in importance due to increasing energy costs and the desire to conserve natural resources. Energy conservation extends into computer systems because of these systems' performance demands, energy costs, and the growth of environmental awareness.\u0000 Meanwhile, computer virtualization is growing in popularity in today's computer systems because it allows for the consolidation of multiple servers onto one larger server. Data centers and computational clusters use virtualization because of the many benefits it offers over the use of traditional stand-alone servers such as ease of management, enhanced security, and reduced costs.\u0000 Virtual machines can be consolidated to obtain maximum power efficiency by taking into account the attributes of applications running in virtual machines. The contribution of this paper is the study of the relationship between applications running in virtual machines, and the effect they have on power consumption. Optimal power efficiency is obtained when virtual machines are consolidated so that all the resources for a given host are fully utilized. This will be used in the goal of building an energy efficient system model and load-balancing algorithm for use in data centers and computational clusters.","PeriodicalId":333104,"journal":{"name":"ACM SE '10","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114329967","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}
As computers progressively dominate our everyday lives, the issue of security becomes increasingly more important. Every electronic account opened has a user name and password attached to it. Usually the password assigned is an arbitrary, random number given by the entity that issues the account. That means the more accounts we have, e.g. e-mail, bank account, office intranets, etc., the more passwords we must remember. Each of these passwords is an alphanumeric sequence. Since it is also recommended that we change passwords every three to six months, it could rapidly become burdensome to remember, not only all the passwords but which password goes with which account. In an effort to simplify the process, many computer users tend to select one generic password and apply it to all accounts. This then becomes a very high security risk since it becomes relatively simple to access all accounts once the first one has been breached. Using the concept that it is easier for the human brain to remember faces than it is to remember alphanumeric sequences; Real User developed an alternative called "Passface". This concept requires users to remember a sequence of five faces as their password rather than alphanumeric characters. Although this may be eminently suitable for the general public, it would be difficult for persons with visual disabilities to use this type of system. The goal of this research effort is to demonstrate that a similar, yet effective, concept can be implemented using sounds instead of faces.
{"title":"Using passtones instead of passwords","authors":"Mary Brown, Felicia R. Doswell","doi":"10.1145/1900008.1900119","DOIUrl":"https://doi.org/10.1145/1900008.1900119","url":null,"abstract":"As computers progressively dominate our everyday lives, the issue of security becomes increasingly more important. Every electronic account opened has a user name and password attached to it. Usually the password assigned is an arbitrary, random number given by the entity that issues the account. That means the more accounts we have, e.g. e-mail, bank account, office intranets, etc., the more passwords we must remember. Each of these passwords is an alphanumeric sequence. Since it is also recommended that we change passwords every three to six months, it could rapidly become burdensome to remember, not only all the passwords but which password goes with which account. In an effort to simplify the process, many computer users tend to select one generic password and apply it to all accounts. This then becomes a very high security risk since it becomes relatively simple to access all accounts once the first one has been breached.\u0000 Using the concept that it is easier for the human brain to remember faces than it is to remember alphanumeric sequences; Real User developed an alternative called \"Passface\". This concept requires users to remember a sequence of five faces as their password rather than alphanumeric characters. Although this may be eminently suitable for the general public, it would be difficult for persons with visual disabilities to use this type of system. The goal of this research effort is to demonstrate that a similar, yet effective, concept can be implemented using sounds instead of faces.","PeriodicalId":333104,"journal":{"name":"ACM SE '10","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128664771","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 continuous research on data analysis based on our previous work on similarity search problems. PanKNN [13] is a novel technique which explores the meaning of K nearest neighbors from a new perspective, redefines the distances between data points and a given query point Q, and efficiently and effectively selects data points which are closest to Q. It can be applied in various data mining fields. In this paper, we present our approach to improving the PanKNN algorithm using the Shrinking concept. Shrinking[15] is a data preprocessing technique which optimizes the inner structure of data inspired by the Newton's Universal Law of Gravitation[11] in the real world. This improved approach can assist to improve the performance of existing data analysis approaches.
{"title":"Towards improving a similarity search approach","authors":"Yong Shi","doi":"10.1145/1900008.1900076","DOIUrl":"https://doi.org/10.1145/1900008.1900076","url":null,"abstract":"In this paper, we present continuous research on data analysis based on our previous work on similarity search problems. PanKNN [13] is a novel technique which explores the meaning of K nearest neighbors from a new perspective, redefines the distances between data points and a given query point Q, and efficiently and effectively selects data points which are closest to Q. It can be applied in various data mining fields. In this paper, we present our approach to improving the PanKNN algorithm using the Shrinking concept. Shrinking[15] is a data preprocessing technique which optimizes the inner structure of data inspired by the Newton's Universal Law of Gravitation[11] in the real world. This improved approach can assist to improve the performance of existing data analysis approaches.","PeriodicalId":333104,"journal":{"name":"ACM SE '10","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129351414","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}
For applications hosted in a cloud computing system, where there are many servers to handle incoming invocations of the application, the assessment of stability of the cloud is very important for both the planning of new applications and the expansion of existing applications. However, a general assessment is always hard to achieve as there are still no standard definitions of techniques used in cloud computing and an actual cloud computing system could be very large and complex. This paper presents a simulation for a cloud computing environment. It enables an assessment of the cloud's logical stability under various configurations without performing experiments on the actual cloud environment. The correctness of the simulation is verified by the theoretical calculation results of the well known M/M/1 queuing system.
{"title":"Using simulations to assess the stability and capacity of cloud computing systems","authors":"Jingsong Wang, M. Huhns","doi":"10.1145/1900008.1900108","DOIUrl":"https://doi.org/10.1145/1900008.1900108","url":null,"abstract":"For applications hosted in a cloud computing system, where there are many servers to handle incoming invocations of the application, the assessment of stability of the cloud is very important for both the planning of new applications and the expansion of existing applications. However, a general assessment is always hard to achieve as there are still no standard definitions of techniques used in cloud computing and an actual cloud computing system could be very large and complex. This paper presents a simulation for a cloud computing environment. It enables an assessment of the cloud's logical stability under various configurations without performing experiments on the actual cloud environment. The correctness of the simulation is verified by the theoretical calculation results of the well known M/M/1 queuing system.","PeriodicalId":333104,"journal":{"name":"ACM SE '10","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121774690","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 Screenplay Scout interface enables a filmmaker to scout camera placements and angles in a virtual 3D world using a lightweight and low-cost combination of a gamepad and a three-degree of freedom (3DOF) orientation sensor. He or she records the camera moves as the orientation sensor's real-time 3DOF data is mapped to the aim direction of the virtual camera and the gamepad controls camera position and the zoom lens. During recording the Screenplay Scout automatically saves to disk still frame images and a text log of camera setting data to support creation of storyboards and cinematic previsualization.
{"title":"Virtual camera scouting with 3DOF motion trackers","authors":"William H. Bares, Nick Buckner, Daniel Castille","doi":"10.1145/1900008.1900129","DOIUrl":"https://doi.org/10.1145/1900008.1900129","url":null,"abstract":"The Screenplay Scout interface enables a filmmaker to scout camera placements and angles in a virtual 3D world using a lightweight and low-cost combination of a gamepad and a three-degree of freedom (3DOF) orientation sensor. He or she records the camera moves as the orientation sensor's real-time 3DOF data is mapped to the aim direction of the virtual camera and the gamepad controls camera position and the zoom lens. During recording the Screenplay Scout automatically saves to disk still frame images and a text log of camera setting data to support creation of storyboards and cinematic previsualization.","PeriodicalId":333104,"journal":{"name":"ACM SE '10","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124772308","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：