Jia Xu, Patrick Shironoshita, U. Visser, N. John, M. Kabuka
The extraction of logically-independent fragments out of an ontology ABox can be useful for solving the tractability problem of querying ontologies with large ABoxes. In this paper, we propose a formal definition of an ABox module, such that it guarantees complete preservation of facts about a given set of individuals, and thus can be reasoned independently w.r.t. the ontology TBox. With ABox modules of this type, isolated or distributed (parallel) ABox reasoning becomes feasible, and more efficient data retrieval from ontology ABoxes can be attained. To compute such an ABox module, we present a theoretical approach and also an approximation for SHIQ ontologies. Evaluation of the module approximation on different types of ontologies shows that, on average, extracted ABox modules are significantly smaller than the entire ABox, and the time for ontology reasoning based on ABox modules can be improved significantly.
{"title":"Module Extraction for Efficient Object Queries over Ontologies with Large ABoxes.","authors":"Jia Xu, Patrick Shironoshita, U. Visser, N. John, M. Kabuka","doi":"10.15764/AIA.2015.01002","DOIUrl":"https://doi.org/10.15764/AIA.2015.01002","url":null,"abstract":"The extraction of logically-independent fragments out of an ontology ABox can be useful for solving the tractability problem of querying ontologies with large ABoxes. In this paper, we propose a formal definition of an ABox module, such that it guarantees complete preservation of facts about a given set of individuals, and thus can be reasoned independently w.r.t. the ontology TBox. With ABox modules of this type, isolated or distributed (parallel) ABox reasoning becomes feasible, and more efficient data retrieval from ontology ABoxes can be attained. To compute such an ABox module, we present a theoretical approach and also an approximation for SHIQ ontologies. Evaluation of the module approximation on different types of ontologies shows that, on average, extracted ABox modules are significantly smaller than the entire ABox, and the time for ontology reasoning based on ABox modules can be improved significantly.","PeriodicalId":91205,"journal":{"name":"Artificial intelligence and applications (Commerce, Calif.)","volume":"59 1","pages":"8-31"},"PeriodicalIF":0.0,"publicationDate":"2015-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74229714","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 problem of revising a belief database is treated in many classical works. We will consider here the problem of merging two belief databases (BDBs for short) Ψ1 and Ψ2, operation that will be denoted by Ψ1 Ψ2, and whose result will be a new BDB. Since belief not necessarily reflects the actual state of the world (as opposed to knowledge), both BDBs could be incompatible. The goal is to construct a new BDB trying to retain as much as possible of the original beliefs of Ψ1 and Ψ2.
{"title":"Merging rule-based belief databases","authors":"R. Wehbe","doi":"10.7892/BORIS.26458","DOIUrl":"https://doi.org/10.7892/BORIS.26458","url":null,"abstract":"The problem of revising a belief database is treated in many classical works. We will consider here the problem of merging two belief databases (BDBs for short) Ψ1 and Ψ2, operation that will be denoted by Ψ1 Ψ2, and whose result will be a new BDB. Since belief not necessarily reflects the actual state of the world (as opposed to knowledge), both BDBs could be incompatible. The goal is to construct a new BDB trying to retain as much as possible of the original beliefs of Ψ1 and Ψ2.","PeriodicalId":91205,"journal":{"name":"Artificial intelligence and applications (Commerce, Calif.)","volume":"40 1","pages":"630-635"},"PeriodicalIF":0.0,"publicationDate":"2007-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78619025","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 concerns with the problem of gathering data from a wireless multi-hop network of energy-constrained sensor nodes to a common base station. To extend lifetime of this network, a certain number of communication nodes are placed among the sensor nodes to relay the acquired data. The main contribution of this work lies in employment of an evolutionary algorithm to determine the best positions of relay nodes in network of randomly placed sensor nodes.
{"title":"Relay Node Placement in Energy-Constrained Networks using SOMA Evolutionary Algorithm","authors":"M. Cervenka, I. Zelinka","doi":"10.5555/1166890.1166897","DOIUrl":"https://doi.org/10.5555/1166890.1166897","url":null,"abstract":"This paper concerns with the problem of gathering data from a wireless multi-hop network of energy-constrained sensor nodes to a common base station. To extend lifetime of this network, a certain number of communication nodes are placed among the sensor nodes to relay the acquired data. The main contribution of this work lies in employment of an evolutionary algorithm to determine the best positions of relay nodes in network of randomly placed sensor nodes.","PeriodicalId":91205,"journal":{"name":"Artificial intelligence and applications (Commerce, Calif.)","volume":"39 14","pages":"40-44"},"PeriodicalIF":0.0,"publicationDate":"2006-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72420188","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}
S. Chimphlee, N. Salim, M. Ngadiman, W. Chimphlee, Surat Srinoy
Web Usage Mining is that area of Web Mining which deals with the extraction of interesting knowledge from logging information produced by Web servers. A challenge in web classification is how to deal with the high dimensionality of the feature space. In this paper we present Independent Component Analysis (ICA) for feature selection and using Rough Fuzzy for clustering web user sessions. It aims at discovery of trends and regularities in web users' access patterns. ICA is a very general-purpose statistical technique in which observed random data are linearly transformed into components that are maximally independent from each other, and simultaneously have "interesting" distributions. Our experiments indicate can improve the predictive performance when the original feature set for representing web log is large and can handling the different groups of uncertainties/impreciseness accuracy.
{"title":"Independent Component Analysis and Rough Fuzzy based Approach to Web Usage Mining","authors":"S. Chimphlee, N. Salim, M. Ngadiman, W. Chimphlee, Surat Srinoy","doi":"10.5555/1166890.1166962","DOIUrl":"https://doi.org/10.5555/1166890.1166962","url":null,"abstract":"Web Usage Mining is that area of Web Mining which deals with the extraction of interesting knowledge from logging information produced by Web servers. A challenge in web classification is how to deal with the high dimensionality of the feature space. In this paper we present Independent Component Analysis (ICA) for feature selection and using Rough Fuzzy for clustering web user sessions. It aims at discovery of trends and regularities in web users' access patterns. ICA is a very general-purpose statistical technique in which observed random data are linearly transformed into components that are maximally independent from each other, and simultaneously have \"interesting\" distributions. Our experiments indicate can improve the predictive performance when the original feature set for representing web log is large and can handling the different groups of uncertainties/impreciseness accuracy.","PeriodicalId":91205,"journal":{"name":"Artificial intelligence and applications (Commerce, Calif.)","volume":"78 1","pages":"422-427"},"PeriodicalIF":0.0,"publicationDate":"2006-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81288988","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}
Representing emotional expressions in text-to-speech synthesis is an interesting subject. The ultimate purpose of our research is to develop an automatic reading system which reads text aloud such as novels with emotion. Our strategy for constructing the system is that we classify the emotion of a text in perspective based on the distribution of emotional words, and classify the emotion of a sentence based on the emotion of nouns, adjectives or verbs composed in the sentence instead of understanding the meaning of the text, and synthesize speech using partially optimum prosodic parameters.
{"title":"A Method for Classifying Emotion of Text based on Emotional Dictionaries for Emotional Reading","authors":"F. Sugimoto, M. Yoneyama","doi":"10.5555/1166890.1166906","DOIUrl":"https://doi.org/10.5555/1166890.1166906","url":null,"abstract":"Representing emotional expressions in text-to-speech synthesis is an interesting subject. The ultimate purpose of our research is to develop an automatic reading system which reads text aloud such as novels with emotion. Our strategy for constructing the system is that we classify the emotion of a text in perspective based on the distribution of emotional words, and classify the emotion of a sentence based on the emotion of nouns, adjectives or verbs composed in the sentence instead of understanding the meaning of the text, and synthesize speech using partially optimum prosodic parameters.","PeriodicalId":91205,"journal":{"name":"Artificial intelligence and applications (Commerce, Calif.)","volume":"149 1","pages":"91-96"},"PeriodicalIF":0.0,"publicationDate":"2006-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78294546","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}
W. Chimphlee, M. Sap, A. Abdullah, S. Chimphlee, Surat Srinoy
The Traditional intrusion detection systems (IDS) look for unusual or suspicious activity, such as patterns of network traffic that are likely indicators of unauthorized activity. However, normal operation often produces traffic that matches likely "attack signature", resulting in false alarms. In this paper we propose an intrusion detection method that proposes rough set based feature selection heuristics and using fuzzy c-means for clustering data. Rough set has to decrease the amount of data and get rid of redundancy. Fuzzy Clustering methods allow objects to belong to several clusters simultaneously, with different degrees of membership. Our approach allows us to recognize not only known attacks but also to increase accuracy detection rate for suspicious activity and signature detection. Empirical studies using the network security data set from the DARPA 1998 offline intrusion detection project (KDD 1999 Cup) show the feasibility of misuse and anomaly detection results.
{"title":"To Identify Suspicious Activity in Anomaly Detection based on Soft Computing","authors":"W. Chimphlee, M. Sap, A. Abdullah, S. Chimphlee, Surat Srinoy","doi":"10.5555/1166890.1166951","DOIUrl":"https://doi.org/10.5555/1166890.1166951","url":null,"abstract":"The Traditional intrusion detection systems (IDS) look for unusual or suspicious activity, such as patterns of network traffic that are likely indicators of unauthorized activity. However, normal operation often produces traffic that matches likely \"attack signature\", resulting in false alarms. In this paper we propose an intrusion detection method that proposes rough set based feature selection heuristics and using fuzzy c-means for clustering data. Rough set has to decrease the amount of data and get rid of redundancy. Fuzzy Clustering methods allow objects to belong to several clusters simultaneously, with different degrees of membership. Our approach allows us to recognize not only known attacks but also to increase accuracy detection rate for suspicious activity and signature detection. Empirical studies using the network security data set from the DARPA 1998 offline intrusion detection project (KDD 1999 Cup) show the feasibility of misuse and anomaly detection results.","PeriodicalId":91205,"journal":{"name":"Artificial intelligence and applications (Commerce, Calif.)","volume":"33 1","pages":"359-364"},"PeriodicalIF":0.0,"publicationDate":"2006-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76696147","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}
A large size of legal knowledge base which consists of entangled inference rules, facts, and arbitrary interpretations may latently include inconsistency within them. In this paper, we propose a method to find the source of such inconsistency by supplying hypothesized facts into a set of rules. With this, we put those rules in the order of reliability and show a stable part of the legal knowledge. First, we define an argument as a chaining of rules to support a certain proposition. Thereafter, we compose a minimal inconsistency set (MIS) combining two disagreeing arguments. Among such a MIS, we can distinguish stable rules that is indifferent to the source of inconsistent from unstable rules, which can be candidates of future amendment. A knowledge-base which consists of stable rules can be also distinguished from that which may contain unstable rules.
{"title":"Stable Legal Knowledge with Regard to Contradictory Arguments","authors":"Shingo Hagiwara, S. Tojo","doi":"10.5555/1166890.1166945","DOIUrl":"https://doi.org/10.5555/1166890.1166945","url":null,"abstract":"A large size of legal knowledge base which consists of entangled inference rules, facts, and arbitrary interpretations may latently include inconsistency within them. In this paper, we propose a method to find the source of such inconsistency by supplying hypothesized facts into a set of rules. With this, we put those rules in the order of reliability and show a stable part of the legal knowledge. First, we define an argument as a chaining of rules to support a certain proposition. Thereafter, we compose a minimal inconsistency set (MIS) combining two disagreeing arguments. Among such a MIS, we can distinguish stable rules that is indifferent to the source of inconsistent from unstable rules, which can be candidates of future amendment. A knowledge-base which consists of stable rules can be also distinguished from that which may contain unstable rules.","PeriodicalId":91205,"journal":{"name":"Artificial intelligence and applications (Commerce, Calif.)","volume":"265 1","pages":"323-328"},"PeriodicalIF":0.0,"publicationDate":"2006-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75917879","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}
Reinforcement learning (RL) is a machine learning technique for sequential decision making. This approach is well proven in many small-scale domains. The true potential of this technique cannot be fully realised until it can adequately deal with the large domain sizes that typically describe real world problems. RL with function approximation is one method of dealing with the domain size problem. This paper investigates two different function approximation approaches to RL: Fuzzy Sarsa and gradient descent Sarsa(λ) with tile coding. It presents detailed experiments in two different simulation environments on the effectiveness of the two approaches. Initial experiments indicated that the tile coding approach had greater modelling capabilities in both testbed domains. However, experimentation in a coevolutionary scenario has indicated that Fuzzy Sarsa has greater flexibility.
{"title":"Fuzzy and Tile Coding Function Approximation in Agent Coevolution","authors":"L. Tokarchuk, J. Bigham, L. Cuthbert","doi":"10.5555/1166890.1166950","DOIUrl":"https://doi.org/10.5555/1166890.1166950","url":null,"abstract":"Reinforcement learning (RL) is a machine learning technique for sequential decision making. This approach is well proven in many small-scale domains. The true potential of this technique cannot be fully realised until it can adequately deal with the large domain sizes that typically describe real world problems. RL with function approximation is one method of dealing with the domain size problem. This paper investigates two different function approximation approaches to RL: Fuzzy Sarsa and gradient descent Sarsa(λ) with tile coding. It presents detailed experiments in two different simulation environments on the effectiveness of the two approaches. Initial experiments indicated that the tile coding approach had greater modelling capabilities in both testbed domains. However, experimentation in a coevolutionary scenario has indicated that Fuzzy Sarsa has greater flexibility.","PeriodicalId":91205,"journal":{"name":"Artificial intelligence and applications (Commerce, Calif.)","volume":"66 1","pages":"353-358"},"PeriodicalIF":0.0,"publicationDate":"2006-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90406127","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 the theory of rough sets of data-mining, a subset of a database represents a certain knowledge. Thus to determine the subset in the database is equivalent to obtain the knowledges which the database possesses. A topological space is constructed by the database. An open subset in the topological space defined by the database corresponds to a certain knowledge in the database. Here we consider topological properties of approximation spaces in generalized rough sets. We show that (a) If R is reflexive and transitive, then R = R (T(R)). Conversely, if R = R(T (R)), then R is reflexive and transitive.(b)If O is a topology with a property (IP), then O = T(R(O)), where (IP) means that Aλ ∈ O(λ ∈ Λ) implies ∩λ Aλ ∈ O. Conversely, for any topology O, if O=T(R(O)), then it satisfies (IP).
在数据挖掘的粗糙集理论中，一个数据库的子集代表一个特定的知识。因此，确定数据库中的子集相当于获得数据库所拥有的知识。由数据库构造拓扑空间。数据库定义的拓扑空间中的开放子集对应于数据库中的某个知识。本文研究广义粗糙集中近似空间的拓扑性质。我们证明了(a)如果R是自反传递的，则R = R (T(R))。反之，若R = R(T (R))，则R是自反可传递的。(b)若O是具有属性(IP)的拓扑，则O=T(R(O))，其中(IP)表示λ∈O(λ∈Λ)暗示∩λ λ∈O。
{"title":"On Topologies Defined by Binary Relations in Rough Sets","authors":"M. Kondo","doi":"10.5555/1166890.1166959","DOIUrl":"https://doi.org/10.5555/1166890.1166959","url":null,"abstract":"In the theory of rough sets of data-mining, a subset of a database represents a certain knowledge. Thus to determine the subset in the database is equivalent to obtain the knowledges which the database possesses. A topological space is constructed by the database. An open subset in the topological space defined by the database corresponds to a certain knowledge in the database. Here we consider topological properties of approximation spaces in generalized rough sets. We show that (a) If R is reflexive and transitive, then R = R (T(R)). Conversely, if R = R(T (R)), then R is reflexive and transitive.(b)If O is a topology with a property (IP), then O = T(R(O)), where (IP) means that Aλ ∈ O(λ ∈ Λ) implies ∩λ Aλ ∈ O. Conversely, for any topology O, if O=T(R(O)), then it satisfies (IP).","PeriodicalId":91205,"journal":{"name":"Artificial intelligence and applications (Commerce, Calif.)","volume":"119 1","pages":"66-77"},"PeriodicalIF":0.0,"publicationDate":"2006-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74288758","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}
While artificial neural networks (NN) promise superior performance in forecasting theory, they are not an established method in business practice. The vast degrees of freedom in modeling NNs have lead to countless publications on heuristic approaches to simplify modeling, training, network selection and evaluation. However, not all studies have conducted experiments with the same scientific rigor, limiting their relevance to further NN research and practice. Consequently, we propose a systematic evaluation to identify successful heuristics and derive sound guidelines to NN modeling from publications. As each forecasting domain of predictive classification or regression imposes different heuristics on specific datasets, a literature review is conducted, identifying 47 publications within the homogeneous business domain of sales forecasting and demand planning out of 4790 publications within the domain of NN forecasting. The identified publications are evaluated through a framework regarding their validity in experiment design and reliability through documentation, in order to identify and promote preeminent publications, derive recommendations for future experiments and identify gaps in current research and practice.
{"title":"An Extended Evaluation Framework for Neural Network Publications in Sales Forecasting","authors":"S. Crone, D. Preßmar","doi":"10.5555/1166890.1166921","DOIUrl":"https://doi.org/10.5555/1166890.1166921","url":null,"abstract":"While artificial neural networks (NN) promise superior performance in forecasting theory, they are not an established method in business practice. The vast degrees of freedom in modeling NNs have lead to countless publications on heuristic approaches to simplify modeling, training, network selection and evaluation. However, not all studies have conducted experiments with the same scientific rigor, limiting their relevance to further NN research and practice. Consequently, we propose a systematic evaluation to identify successful heuristics and derive sound guidelines to NN modeling from publications. As each forecasting domain of predictive classification or regression imposes different heuristics on specific datasets, a literature review is conducted, identifying 47 publications within the homogeneous business domain of sales forecasting and demand planning out of 4790 publications within the domain of NN forecasting. The identified publications are evaluated through a framework regarding their validity in experiment design and reliability through documentation, in order to identify and promote preeminent publications, derive recommendations for future experiments and identify gaps in current research and practice.","PeriodicalId":91205,"journal":{"name":"Artificial intelligence and applications (Commerce, Calif.)","volume":"51 1","pages":"179-186"},"PeriodicalIF":0.0,"publicationDate":"2006-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80963120","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：