Pub Date : 2018-09-01DOI: 10.1109/SYNASC.2018.00070
A. Enescu, A. Andreica, L. Dioşan
This paper presents an edge detection method based on Cellular Automata where the rules are evolved to optimize the edge detection in binary images. This method divides the edge detection problem into two sub–problems: on the one hand it trains the rules to detect the edge pixels, on the other hand it trains the rules to detect non–edge (background) pixels. Two best packets of rules are obtained from the training process. These packets of rules are applied in different orders or after they have been processed, thus resulting four different images on which the detection performance of the proposed method is evaluated.
{"title":"Evolving Cellular Automata for Two-Stage Edge Detection","authors":"A. Enescu, A. Andreica, L. Dioşan","doi":"10.1109/SYNASC.2018.00070","DOIUrl":"https://doi.org/10.1109/SYNASC.2018.00070","url":null,"abstract":"This paper presents an edge detection method based on Cellular Automata where the rules are evolved to optimize the edge detection in binary images. This method divides the edge detection problem into two sub–problems: on the one hand it trains the rules to detect the edge pixels, on the other hand it trains the rules to detect non–edge (background) pixels. Two best packets of rules are obtained from the training process. These packets of rules are applied in different orders or after they have been processed, thus resulting four different images on which the detection performance of the proposed method is evaluated.","PeriodicalId":273805,"journal":{"name":"2018 20th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127305529","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 : 2018-09-01DOI: 10.1109/SYNASC.2018.00056
Mihaela Ilie, S. Ilie, Ionuţ Murareţu
In our previous work we have introduced a skill-based mathematical model of resource allocation. This paper extends our skill based approach by introducing adaptive skill sets for employees and a history-based initial evaluation strategy. For this purpose, the mathematical model is adjusted in order to modify skill vectors after a task allocation. In turn this enables estimations of the time to task completion based on employee history. This approach would provide the team leaders with a better view of the skill sets mastered by the team. We experimentally evaluate the impact of the skill adjustment on the project duration and cost in a simulation environment. The conclusion of the experiment is that taking into account the implicit skill gain of employees during their daily activity decreases projected costs and execution time significantly, which is this paper's contribution to the state of the art. This approach is a good way to keep the team's skill sets automatically updated. The experiment is designed as an agent society simulation and through their interactions raw data is collected in order to calculate the performance measures. A scalability experiment is also presented showing slight (1%) decreases in project duration when the task number doubles while costs decrease between 7-32%.
{"title":"An Adaptive Recommender System for Human Resource Allocation in Software Projects - Initial Results on an Agent-Based Simulation","authors":"Mihaela Ilie, S. Ilie, Ionuţ Murareţu","doi":"10.1109/SYNASC.2018.00056","DOIUrl":"https://doi.org/10.1109/SYNASC.2018.00056","url":null,"abstract":"In our previous work we have introduced a skill-based mathematical model of resource allocation. This paper extends our skill based approach by introducing adaptive skill sets for employees and a history-based initial evaluation strategy. For this purpose, the mathematical model is adjusted in order to modify skill vectors after a task allocation. In turn this enables estimations of the time to task completion based on employee history. This approach would provide the team leaders with a better view of the skill sets mastered by the team. We experimentally evaluate the impact of the skill adjustment on the project duration and cost in a simulation environment. The conclusion of the experiment is that taking into account the implicit skill gain of employees during their daily activity decreases projected costs and execution time significantly, which is this paper's contribution to the state of the art. This approach is a good way to keep the team's skill sets automatically updated. The experiment is designed as an agent society simulation and through their interactions raw data is collected in order to calculate the performance measures. A scalability experiment is also presented showing slight (1%) decreases in project duration when the task number doubles while costs decrease between 7-32%.","PeriodicalId":273805,"journal":{"name":"2018 20th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128896118","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 : 2018-09-01DOI: 10.1109/SYNASC.2018.00063
Ciprian-Petrisor Pungila, Darius Galis, V. Negru
Pattern-matching techniques are very common in major areas of bioinformatics, in multiple forms: from exact (accurate) to partial matching, the process itself is vital to multiple niches of research. In this paper, we prepare a survey of recent breakthroughs in the field of pattern-matching applied to bioinformatics, from a heterogeneous implementation standpoint, focusing especially on the ones based on SIMD (Single Instruction Multiple Data) or GPGPU (General-Purpose computing on Graphics Processing Units) architectures. We focus on the most important aspects of such data processing and their effectiveness, with particular focus on the technological challenges that such heterogeneous implementations bring, while also analyzing their feasibility of application to particular research niches, such as DNA analysis and protein sequence alignment.
模式匹配技术在生物信息学的主要领域非常常见，有多种形式：从精确（准确）匹配到部分匹配，这一过程本身对多个研究领域至关重要。在本文中，我们从异构实现的角度出发，对应用于生物信息学的模式匹配领域的最新突破进行了调查，尤其侧重于基于 SIMD（单指令多数据）或 GPGPU（图形处理单元上的通用计算）架构的模式匹配。我们将重点放在此类数据处理的最重要方面及其有效性上，尤其关注此类异构实现所带来的技术挑战，同时分析它们应用于特定研究领域（如 DNA 分析和蛋白质序列比对）的可行性。
{"title":"Survey on Feasibility of Pattern Matching Techniques In Heterogeneous Architectures for Bioinformatics","authors":"Ciprian-Petrisor Pungila, Darius Galis, V. Negru","doi":"10.1109/SYNASC.2018.00063","DOIUrl":"https://doi.org/10.1109/SYNASC.2018.00063","url":null,"abstract":"Pattern-matching techniques are very common in major areas of bioinformatics, in multiple forms: from exact (accurate) to partial matching, the process itself is vital to multiple niches of research. In this paper, we prepare a survey of recent breakthroughs in the field of pattern-matching applied to bioinformatics, from a heterogeneous implementation standpoint, focusing especially on the ones based on SIMD (Single Instruction Multiple Data) or GPGPU (General-Purpose computing on Graphics Processing Units) architectures. We focus on the most important aspects of such data processing and their effectiveness, with particular focus on the technological challenges that such heterogeneous implementations bring, while also analyzing their feasibility of application to particular research niches, such as DNA analysis and protein sequence alignment.","PeriodicalId":273805,"journal":{"name":"2018 20th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128123702","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 : 2018-09-01DOI: 10.1109/SYNASC.2018.00017
Enachescu Denis, Enachescu Cornelia
Bayesian networks, BN, are a formalism for probabilistic reasoning that have grown increasingly popular for tasks such as classification in data-mining. In some situations, the structure of the Bayesian network can be given by an expert. If not, retrieving it automatically from a database of cases is a NP-hard problem; notably because of the complexity of the search space. In the last decade, numerous methods have been introduced to learn the networks structure automatically, by simplifying the search space or by using a heuristic in the search space. Most methods deal with completely observed data, but some can deal with incomplete data. In this tutorial we will present, besides BN, other popular classification methods, i.e. Multilayer Perceptrons Network (MLP) and K-nearest neighbor (KNN) an analyze their performance in the context of medical diagnosis.
{"title":"Inferring, Learning and Modelling Complex Systems with Bayesian Networks. A Tutorial","authors":"Enachescu Denis, Enachescu Cornelia","doi":"10.1109/SYNASC.2018.00017","DOIUrl":"https://doi.org/10.1109/SYNASC.2018.00017","url":null,"abstract":"Bayesian networks, BN, are a formalism for probabilistic reasoning that have grown increasingly popular for tasks such as classification in data-mining. In some situations, the structure of the Bayesian network can be given by an expert. If not, retrieving it automatically from a database of cases is a NP-hard problem; notably because of the complexity of the search space. In the last decade, numerous methods have been introduced to learn the networks structure automatically, by simplifying the search space or by using a heuristic in the search space. Most methods deal with completely observed data, but some can deal with incomplete data. In this tutorial we will present, besides BN, other popular classification methods, i.e. Multilayer Perceptrons Network (MLP) and K-nearest neighbor (KNN) an analyze their performance in the context of medical diagnosis.","PeriodicalId":273805,"journal":{"name":"2018 20th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114074504","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 : 2018-09-01DOI: 10.1109/SYNASC.2018.00054
E. Todoran
In recent work we have introduced an experimental concurrent programming language which supports a systematic approach to performance analysis and formal verification correlated with a programming style called performance evaluation programming [19]. For the purpose of formal verification, the ranges of variables must be bounded and concurrent programs are translated into corresponding (finite state) Continuous Time Markov Chains (CTMCs) which are analyzed by using the PRISM tool. Activities in a CTMC model are abstracted by their rates. In the language introduced in [19] an activity is the evaluation of a function expressed in a functional sub-language. The solution presented in [19] supports formal verification in a systematic manner, but not automatically, requiring the programmer to generate certain data for the performance evaluation experiments. In this paper we refine the design of the functional sub-language introduced in [19] by using concepts of functional programming with dependent types. We use dependent types to control the ranges of variables. The solution presented in this paper is devised to support automatic performance evaluation and formal verification of (bounded versions of) concurrent programs.
{"title":"Towards Performance Evaluation Programming","authors":"E. Todoran","doi":"10.1109/SYNASC.2018.00054","DOIUrl":"https://doi.org/10.1109/SYNASC.2018.00054","url":null,"abstract":"In recent work we have introduced an experimental concurrent programming language which supports a systematic approach to performance analysis and formal verification correlated with a programming style called performance evaluation programming [19]. For the purpose of formal verification, the ranges of variables must be bounded and concurrent programs are translated into corresponding (finite state) Continuous Time Markov Chains (CTMCs) which are analyzed by using the PRISM tool. Activities in a CTMC model are abstracted by their rates. In the language introduced in [19] an activity is the evaluation of a function expressed in a functional sub-language. The solution presented in [19] supports formal verification in a systematic manner, but not automatically, requiring the programmer to generate certain data for the performance evaluation experiments. In this paper we refine the design of the functional sub-language introduced in [19] by using concepts of functional programming with dependent types. We use dependent types to control the ranges of variables. The solution presented in this paper is devised to support automatic performance evaluation and formal verification of (bounded versions of) concurrent programs.","PeriodicalId":273805,"journal":{"name":"2018 20th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124318295","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 : 2018-09-01DOI: 10.1109/SYNASC.2018.00066
D. Lucanu
The coinduction is dual to induction and both of them can be defined as fixed points. More precisely, a set is inductive if it is the least fixed-point (lfp) of a monotone endofunction on a complete lattice, and it is coinductive if it is the greatest fixed-point (gfp) of such a endofunction. The induction principle says that each set that is a pre-fixed point includes the lfp, and the coinduction principle says that any post-fixed point is included in the gfp. A convenient way to define (co) inductive sets is by means of rules. In contrast with the induction, which is a well-known proof principle that is taught in most undergraduate programs, the coinduction is not as widespread and its main applications includes bisimulation and behavioural equivalence. In this talk we show that the reachability properties of transition systems can be defined coinductively and we present coinductive proof systems for such properties, where the transition systems are specified by Logically constrained term rewriting systems. As an application, if the transition system describes the semantics of a programming language, then the reachability properties may be used to describe the partial correctness of programs. A main advantage of the presented proof system is that it can be automated. Logically constrained term rewriting systems are parametric in a builtin model, for which an automated theorem prover (e.g. a SMT solver) cand be used.
{"title":"Proving Reachability Properties by Coinduction (Extended Abstract)","authors":"D. Lucanu","doi":"10.1109/SYNASC.2018.00066","DOIUrl":"https://doi.org/10.1109/SYNASC.2018.00066","url":null,"abstract":"The coinduction is dual to induction and both of them can be defined as fixed points. More precisely, a set is inductive if it is the least fixed-point (lfp) of a monotone endofunction on a complete lattice, and it is coinductive if it is the greatest fixed-point (gfp) of such a endofunction. The induction principle says that each set that is a pre-fixed point includes the lfp, and the coinduction principle says that any post-fixed point is included in the gfp. A convenient way to define (co) inductive sets is by means of rules. In contrast with the induction, which is a well-known proof principle that is taught in most undergraduate programs, the coinduction is not as widespread and its main applications includes bisimulation and behavioural equivalence. In this talk we show that the reachability properties of transition systems can be defined coinductively and we present coinductive proof systems for such properties, where the transition systems are specified by Logically constrained term rewriting systems. As an application, if the transition system describes the semantics of a programming language, then the reachability properties may be used to describe the partial correctness of programs. A main advantage of the presented proof system is that it can be automated. Logically constrained term rewriting systems are parametric in a builtin model, for which an automated theorem prover (e.g. a SMT solver) cand be used.","PeriodicalId":273805,"journal":{"name":"2018 20th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115638174","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 : 2018-09-01DOI: 10.1109/SYNASC.2018.00072
Georgiana Ingrid Stoleru, Adrian-Stefan Popescu, Dragos Gavrilut
Cyberattacks have evolved from infecting computers using floppy disks or USB drives to the point where Internet, through malicious URLs or spear phishing, has become the main infection vector. In order for these attacks to succeed and avoid detection, an attacker must often change the location where the malicious content is hosted. The short life span of a malicious URL has forced many security vendors to search for different proactive methods for detection. Therefore, machine learning algorithms have become a powerful tool against this kind of attack vectors. The paper presents multiple approaches to combine features obtained from URL body and from its content in order to increase the detection rate for Internet attacks, taking into consideration the short life span of malicious URLs and the high importance of keeping the false positives rate to a minimum.
{"title":"Increasing Protection Against Internet Attacks through Contextual Feature Pairing","authors":"Georgiana Ingrid Stoleru, Adrian-Stefan Popescu, Dragos Gavrilut","doi":"10.1109/SYNASC.2018.00072","DOIUrl":"https://doi.org/10.1109/SYNASC.2018.00072","url":null,"abstract":"Cyberattacks have evolved from infecting computers using floppy disks or USB drives to the point where Internet, through malicious URLs or spear phishing, has become the main infection vector. In order for these attacks to succeed and avoid detection, an attacker must often change the location where the malicious content is hosted. The short life span of a malicious URL has forced many security vendors to search for different proactive methods for detection. Therefore, machine learning algorithms have become a powerful tool against this kind of attack vectors. The paper presents multiple approaches to combine features obtained from URL body and from its content in order to increase the detection rate for Internet attacks, taking into consideration the short life span of malicious URLs and the high importance of keeping the false positives rate to a minimum.","PeriodicalId":273805,"journal":{"name":"2018 20th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122915279","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 : 2018-09-01DOI: 10.1109/SYNASC.2018.00061
D. Rotar, H. Andreescu
Already available Advanced driver-assistance systems technologies, like lane departure, cruse control, blind spot detection, etc. are constructing a model of the external environment, helping to increase the security in traffic. With the advance of the autonomous driving, the car makers are shifting the focus also into the interior of the car. Face recognition, i.e. recognizing a person, based on a picture, which is highly available on social platforms(e.g. Facebook) poses a challenge when it has to be implemented in an automotive system, due to smaller computational power and lower memory currently available in such system. This paper explores the possibility to implement a face recognition on an automotive embedded system. The system is composed from a infrared camera, 1MPixel resolution, and an image processing unit based on TDA3x
{"title":"Face Recognition in Automotive Applications","authors":"D. Rotar, H. Andreescu","doi":"10.1109/SYNASC.2018.00061","DOIUrl":"https://doi.org/10.1109/SYNASC.2018.00061","url":null,"abstract":"Already available Advanced driver-assistance systems technologies, like lane departure, cruse control, blind spot detection, etc. are constructing a model of the external environment, helping to increase the security in traffic. With the advance of the autonomous driving, the car makers are shifting the focus also into the interior of the car. Face recognition, i.e. recognizing a person, based on a picture, which is highly available on social platforms(e.g. Facebook) poses a challenge when it has to be implemented in an automotive system, due to smaller computational power and lower memory currently available in such system. This paper explores the possibility to implement a face recognition on an automotive embedded system. The system is composed from a infrared camera, 1MPixel resolution, and an image processing unit based on TDA3x","PeriodicalId":273805,"journal":{"name":"2018 20th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124955498","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 : 2018-09-01DOI: 10.1109/SYNASC.2018.00044
Z. Bodó
Data cleaning constitutes an important problem in information science. Collecting data about the same entities from multiple sources or following distinct methodologies might result in slightly different, inconsistent data. The objective of data cleaning is to produce a fused version combining the differing data, resulting in a cleaner dataset. In this paper we collect document metadata records from CiteSeerX and build a supervised record linker to Crossref. The supervised method is trained using a manually linked dataset containing 512 verified DOIs—to our knowledge, up to now being the largest such dataset for bibliographic record linkage. We experiment using different supervised learning methods, and also prove experimentally that the accuracy of the attached metadata records can improve the performance of automatic metadata extraction systems.
{"title":"A CiteSeerX-Based Dataset for Record Linkage and Metadata Extraction","authors":"Z. Bodó","doi":"10.1109/SYNASC.2018.00044","DOIUrl":"https://doi.org/10.1109/SYNASC.2018.00044","url":null,"abstract":"Data cleaning constitutes an important problem in information science. Collecting data about the same entities from multiple sources or following distinct methodologies might result in slightly different, inconsistent data. The objective of data cleaning is to produce a fused version combining the differing data, resulting in a cleaner dataset. In this paper we collect document metadata records from CiteSeerX and build a supervised record linker to Crossref. The supervised method is trained using a manually linked dataset containing 512 verified DOIs—to our knowledge, up to now being the largest such dataset for bibliographic record linkage. We experiment using different supervised learning methods, and also prove experimentally that the accuracy of the attached metadata records can improve the performance of automatic metadata extraction systems.","PeriodicalId":273805,"journal":{"name":"2018 20th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121609512","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 : 2018-09-01DOI: 10.1109/SYNASC.2018.00055
S. Szabó, Bogdán Zaválnij
The way we tackle NP-hard problems in practical setting has experienced a major shift in recent years. Our view has became more sophisticated with the emergence of the parameterized complexity paradigm. We may distinguish subclasses inside the NP-hard complexity class. The complexity of the problems in different subclasses maybe quite different. The overall conservative estimate of the running time is replaced by a more optimistic estimate. In addition the approach of parameterized algorithms is sometimes able to deal with the more complex problems by dividing the problem into harder and a simpler parts. The easier instance at many times reduces to mere preprocessing step leaving us with only the harder part. In this paper we single out the so-called maximum clique problem as a typical representative of the NP-hard complexity class. We propose an algorithm to solve the maximum clique problem motivated by the above ideas. Many of the available maximum clique solvers are descendants or refined versions of the Carraghan–Pardalos algorithm. (Patric "Ostergaa rd's cliquer is being as an exception.) The maximum clique problem as a maximization problem can be reduced to a series of k-clique problems as decision problems. Our main observation is that this route offers a number of advantages. The structure of a k-clique decision problem is simpler than the structure of a maximization problem. It affords additional pruning opportunities based on the available value of k. A large scale numerical experiment indicates that in many occasions the combined search space of the k-clique problems is smaller than the search space of the maximization problem. The solver we propose turns out to be rather efficient. In a number of test problems it beats the best available solvers.
{"title":"A Different Approach to Maximum Clique Search","authors":"S. Szabó, Bogdán Zaválnij","doi":"10.1109/SYNASC.2018.00055","DOIUrl":"https://doi.org/10.1109/SYNASC.2018.00055","url":null,"abstract":"The way we tackle NP-hard problems in practical setting has experienced a major shift in recent years. Our view has became more sophisticated with the emergence of the parameterized complexity paradigm. We may distinguish subclasses inside the NP-hard complexity class. The complexity of the problems in different subclasses maybe quite different. The overall conservative estimate of the running time is replaced by a more optimistic estimate. In addition the approach of parameterized algorithms is sometimes able to deal with the more complex problems by dividing the problem into harder and a simpler parts. The easier instance at many times reduces to mere preprocessing step leaving us with only the harder part. In this paper we single out the so-called maximum clique problem as a typical representative of the NP-hard complexity class. We propose an algorithm to solve the maximum clique problem motivated by the above ideas. Many of the available maximum clique solvers are descendants or refined versions of the Carraghan–Pardalos algorithm. (Patric \"Ostergaa rd's cliquer is being as an exception.) The maximum clique problem as a maximization problem can be reduced to a series of k-clique problems as decision problems. Our main observation is that this route offers a number of advantages. The structure of a k-clique decision problem is simpler than the structure of a maximization problem. It affords additional pruning opportunities based on the available value of k. A large scale numerical experiment indicates that in many occasions the combined search space of the k-clique problems is smaller than the search space of the maximization problem. The solver we propose turns out to be rather efficient. In a number of test problems it beats the best available solvers.","PeriodicalId":273805,"journal":{"name":"2018 20th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126264745","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：