Patricia J. Crossno, Andrew T. Wilson, Timothy M. Shead, Daniel M. Dunlavy
We present Topic View, an application for visually comparing and exploring multiple models of text corpora. Topic View uses multiple linked views to visually analyze both the conceptual content and the document relationships in models generated using different algorithms. To illustrate Topic View, we apply it to models created using two standard approaches: Latent Semantic Analysis (LSA) and Latent Dirichlet Allocation (LDA). Conceptual content is compared through the combination of (i) a bipartite graph matching LSA concepts with LDA topics based on the cosine similarities of model factors and (ii) a table containing the terms for each LSA concept and LDA topic listed in decreasing order of importance. Document relationships are examined through the combination of (i) side-by-side document similarity graphs, (ii) a table listing the weights for each document's contribution to each concept/topic, and (iii) a full text reader for documents selected in either of the graphs or the table. We demonstrate the utility of Topic View's visual approach to model assessment by comparing LSA and LDA models of two example corpora.
{"title":"TopicView: Visually Comparing Topic Models of Text Collections","authors":"Patricia J. Crossno, Andrew T. Wilson, Timothy M. Shead, Daniel M. Dunlavy","doi":"10.1109/ICTAI.2011.162","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.162","url":null,"abstract":"We present Topic View, an application for visually comparing and exploring multiple models of text corpora. Topic View uses multiple linked views to visually analyze both the conceptual content and the document relationships in models generated using different algorithms. To illustrate Topic View, we apply it to models created using two standard approaches: Latent Semantic Analysis (LSA) and Latent Dirichlet Allocation (LDA). Conceptual content is compared through the combination of (i) a bipartite graph matching LSA concepts with LDA topics based on the cosine similarities of model factors and (ii) a table containing the terms for each LSA concept and LDA topic listed in decreasing order of importance. Document relationships are examined through the combination of (i) side-by-side document similarity graphs, (ii) a table listing the weights for each document's contribution to each concept/topic, and (iii) a full text reader for documents selected in either of the graphs or the table. We demonstrate the utility of Topic View's visual approach to model assessment by comparing LSA and LDA models of two example corpora.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124806009","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}
Recently, Semi-Supervised learning algorithms such as co-training are used in many domains. In co-training, two classifiers based on different subsets of the features or on different learning algorithms are trained in parallel and unlabeled data that are classified differently by the classifiers but for which one classifier has large confidence are labeled and used as training data for the other. In this paper, a new form of co-training, called Ensemble-Co-Training, is proposed that uses an ensemble of different learning algorithms. Based on a theorem by Angluin and Laird that relates noise in the data to the error of hypotheses learned from these data, we propose a criterion for finding a subset of high-confidence predictions and error rate for a classifier in each iteration of the training process. Experiments show that the new method in almost all domains gives better results than the state-of-the-art methods.
{"title":"Disagreement-Based Co-training","authors":"J. Tanha, M. Someren, H. Afsarmanesh","doi":"10.1109/ICTAI.2011.126","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.126","url":null,"abstract":"Recently, Semi-Supervised learning algorithms such as co-training are used in many domains. In co-training, two classifiers based on different subsets of the features or on different learning algorithms are trained in parallel and unlabeled data that are classified differently by the classifiers but for which one classifier has large confidence are labeled and used as training data for the other. In this paper, a new form of co-training, called Ensemble-Co-Training, is proposed that uses an ensemble of different learning algorithms. Based on a theorem by Angluin and Laird that relates noise in the data to the error of hypotheses learned from these data, we propose a criterion for finding a subset of high-confidence predictions and error rate for a classifier in each iteration of the training process. Experiments show that the new method in almost all domains gives better results than the state-of-the-art methods.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"287 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124154208","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}
T. Arredondo, P. Castillo, Pablo Benapres, Javiera Quiroz, M. Torres
We describe a biologically inspired memory in a multi-agent based robotic architecture. In this approach, memory and pattern recognition are intertwined to form a cognitive memory that is used for recognition of objects in a robotics environment. This memory is implemented in a multiple agent behavior based blackboard architecture as an object recognition agent. The agent performance is tested against a standard dataset with satisfactory results. The system is currently installed in a mobile robotic platform where its capabilities and applications are explored.
{"title":"A Biologically Inspired Memory in a Multi-agent Based Robotic Architecture","authors":"T. Arredondo, P. Castillo, Pablo Benapres, Javiera Quiroz, M. Torres","doi":"10.1109/ICTAI.2011.58","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.58","url":null,"abstract":"We describe a biologically inspired memory in a multi-agent based robotic architecture. In this approach, memory and pattern recognition are intertwined to form a cognitive memory that is used for recognition of objects in a robotics environment. This memory is implemented in a multiple agent behavior based blackboard architecture as an object recognition agent. The agent performance is tested against a standard dataset with satisfactory results. The system is currently installed in a mobile robotic platform where its capabilities and applications are explored.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124386081","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}
Michael Morak, Nysret Musliu, R. Pichler, Stefan Rümmele, S. Woltran
A promising approach to tackle intractable problems is given by combining decomposition methods with dynamic programming algorithms. One such decomposition concept is tree decomposition. In this paper, we provide a new algorithm using this combined approach for solving reasoning problems in propositional answer set programming.
{"title":"A New Tree-Decomposition Based Algorithm for Answer Set Programming","authors":"Michael Morak, Nysret Musliu, R. Pichler, Stefan Rümmele, S. Woltran","doi":"10.1109/ICTAI.2011.154","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.154","url":null,"abstract":"A promising approach to tackle intractable problems is given by combining decomposition methods with dynamic programming algorithms. One such decomposition concept is tree decomposition. In this paper, we provide a new algorithm using this combined approach for solving reasoning problems in propositional answer set programming.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122579776","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}
Cardinality constraints appear in many practical problems and have been well studied in the past. There are many CNF encodings for cardinality constraints, although it is not clear which encodings perform better. Indeed, different encodings can perform well over different problems. This paper examines a large number of cardinality encodings and evaluates their performance for solving the problem of Maximum Satisfiability (MaxSAT). Taking advantage of the diversification of cardinality encodings, we propose to exploit those encodings in parallel MaxSAT solving. Our parallel solver, pMAX, simultaneously searches in the lower and upper bound of the optimum value, and different cardinality encodings are used in each thread to increase the diversification of the search. Moreover, learned clauses are shared between threads during the search. Experimental results show that our parallel solver outperforms other sequential and parallel state-of-the-art MaxSAT solvers.
{"title":"Exploiting Cardinality Encodings in Parallel Maximum Satisfiability","authors":"R. Martins, Vasco M. Manquinho, I. Lynce","doi":"10.1109/ICTAI.2011.54","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.54","url":null,"abstract":"Cardinality constraints appear in many practical problems and have been well studied in the past. There are many CNF encodings for cardinality constraints, although it is not clear which encodings perform better. Indeed, different encodings can perform well over different problems. This paper examines a large number of cardinality encodings and evaluates their performance for solving the problem of Maximum Satisfiability (MaxSAT). Taking advantage of the diversification of cardinality encodings, we propose to exploit those encodings in parallel MaxSAT solving. Our parallel solver, pMAX, simultaneously searches in the lower and upper bound of the optimum value, and different cardinality encodings are used in each thread to increase the diversification of the search. Moreover, learned clauses are shared between threads during the search. Experimental results show that our parallel solver outperforms other sequential and parallel state-of-the-art MaxSAT solvers.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"147 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122053067","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 study of community detection has received more and more attention in recent years, the problem is very difficult and of great importance in many fields such as sociology, biology and computer science. But most of the algorithms proposed so far could not utilize the weight information within weighted networks, and many of them are so time-consuming that they are not fit for the large-scale networks. We propose a new center-based method, which is especially designed for weighted networks. And the method is also suitable for large-scale network because of its low computational complexity. We demonstrate our method on a synthetic network and two real-world networks. The result shows the high efficiency and precision of our method.
{"title":"A Center-Based Community Detection Method in Weighted Networks","authors":"Jie Jin, Lei Pan, Chong-Jun Wang, Junyuan Xie","doi":"10.1109/ICTAI.2011.83","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.83","url":null,"abstract":"The study of community detection has received more and more attention in recent years, the problem is very difficult and of great importance in many fields such as sociology, biology and computer science. But most of the algorithms proposed so far could not utilize the weight information within weighted networks, and many of them are so time-consuming that they are not fit for the large-scale networks. We propose a new center-based method, which is especially designed for weighted networks. And the method is also suitable for large-scale network because of its low computational complexity. We demonstrate our method on a synthetic network and two real-world networks. The result shows the high efficiency and precision of our method.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123750008","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}
K. Iwata, Toyoshiro Nakashima, Yoshiyuki Anan, N. Ishii
In this paper, we create effort prediction models using self-organizing maps (SOMs) for embedded software development projects. SOMs are a type of artificial neural networks that rely on unsupervised learning. They produce a low-dimensional, discretized representation of the input space of training samples, these representations are called maps. SOMs are useful for visualizing low-dimensional views of high-dimensional data a multidimensional scaling technique. The advantages of using SOMs for statistical applications are as follows: (1) enabling reasonable inferences to be made from incomplete information via association and recollection, (2) visualizing data, (3) summarizing large-scale data, and (4) creating nonlinear models. We focus on the first advantage to create effort prediction models. To verify our approach, we perform an evaluation experiment that compares SOM models to feed forward artificial neural network (FANN) models using Welch's t test. The results of the comparison indicate that SOM models are more accurate than FANN models for the mean of absolute errors when predicting the amount of effort, because mean errors of the SOM are statistically significantly lower.
{"title":"Effort Prediction Models Using Self-Organizing Maps for Embedded Software Development Projects","authors":"K. Iwata, Toyoshiro Nakashima, Yoshiyuki Anan, N. Ishii","doi":"10.1109/ICTAI.2011.30","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.30","url":null,"abstract":"In this paper, we create effort prediction models using self-organizing maps (SOMs) for embedded software development projects. SOMs are a type of artificial neural networks that rely on unsupervised learning. They produce a low-dimensional, discretized representation of the input space of training samples, these representations are called maps. SOMs are useful for visualizing low-dimensional views of high-dimensional data a multidimensional scaling technique. The advantages of using SOMs for statistical applications are as follows: (1) enabling reasonable inferences to be made from incomplete information via association and recollection, (2) visualizing data, (3) summarizing large-scale data, and (4) creating nonlinear models. We focus on the first advantage to create effort prediction models. To verify our approach, we perform an evaluation experiment that compares SOM models to feed forward artificial neural network (FANN) models using Welch's t test. The results of the comparison indicate that SOM models are more accurate than FANN models for the mean of absolute errors when predicting the amount of effort, because mean errors of the SOM are statistically significantly lower.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131730806","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 a world where an increasing number of transactions are made on the web, there is a need for a trust evaluation tool dealing with uncertainty, e.g., for customers interested in evaluating the trustworthiness of an unknown service provider throughout queries to other customers of unknown reliability. In this paper, we propose to estimate the trust of an unknown agent, say ??^D, through the information given by a group of agents who have interacted with agent ??^D. This group of agents is assumed to have an unknown reliability. In order to tackle the uncertainty associated with the trust of unknown agents, we suggest to use possibility distributions. We introduce a new certainty metric to measure the degree of agreement of the information reported by the group of agents about agent ??^D. Fusion rules are then used to estimate the possibility distribution of agent a^D's trust. To the best of our knowledge, this is the first paper that estimates trust, out of empirical data, subject to some uncertainty, in a discrete multi-valued trust environment. Numerical experiments are presented to validate the proposed tools.
{"title":"Under Uncertainty Trust Estimation through Unknown Agents, in a Multi-valued Trust Environment","authors":"Sina Honari, B. Jaumard, J. Bentahar","doi":"10.1109/ICTAI.2011.57","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.57","url":null,"abstract":"In a world where an increasing number of transactions are made on the web, there is a need for a trust evaluation tool dealing with uncertainty, e.g., for customers interested in evaluating the trustworthiness of an unknown service provider throughout queries to other customers of unknown reliability. In this paper, we propose to estimate the trust of an unknown agent, say ??^D, through the information given by a group of agents who have interacted with agent ??^D. This group of agents is assumed to have an unknown reliability. In order to tackle the uncertainty associated with the trust of unknown agents, we suggest to use possibility distributions. We introduce a new certainty metric to measure the degree of agreement of the information reported by the group of agents about agent ??^D. Fusion rules are then used to estimate the possibility distribution of agent a^D's trust. To the best of our knowledge, this is the first paper that estimates trust, out of empirical data, subject to some uncertainty, in a discrete multi-valued trust environment. Numerical experiments are presented to validate the proposed tools.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130480227","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 proposes a model for image representation and image analysis using a multi-layer neural network, which is rooted in the human vision system. Having complex neural layers to represent and process information, the biological vision system is far more efficient than machine vision system. The neural model simulate non-classical receptive field of ganglion cell and its local feedback control circuit, and can represent images, beyond pixel level, self-adaptively and regularly. The results of experiments, rebuilding, distribution and contour detection, prove this method can represent image faithfully with low cost, and can produce a compact and abstract approximation to facilitate successive image segmentation and integration. This representation schema is good at extracting spatial relationships from different components of images and highlighting foreground objects from background, especially for nature images with complicated scenes. Further it can be applied to object recognition or image classification tasks in future.
{"title":"A Bio-inspired Model for Image Representation and Image Analysis","authors":"Hui Wei, Qingsong Zuo, B. Lang","doi":"10.1109/ICTAI.2011.67","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.67","url":null,"abstract":"This paper proposes a model for image representation and image analysis using a multi-layer neural network, which is rooted in the human vision system. Having complex neural layers to represent and process information, the biological vision system is far more efficient than machine vision system. The neural model simulate non-classical receptive field of ganglion cell and its local feedback control circuit, and can represent images, beyond pixel level, self-adaptively and regularly. The results of experiments, rebuilding, distribution and contour detection, prove this method can represent image faithfully with low cost, and can produce a compact and abstract approximation to facilitate successive image segmentation and integration. This representation schema is good at extracting spatial relationships from different components of images and highlighting foreground objects from background, especially for nature images with complicated scenes. Further it can be applied to object recognition or image classification tasks in future.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122200705","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 kind of strategy decision-making risk analysis method is put forward based on uncertain event analysis, on the basis of potential uncertain events confronted in decision-makings. In this method, a model is constructed to generate the decision-making option space, with detailed generation ways based on the risks of uncertain events. By comparing the potential capacities of the enemy, self-army and the third party, the occurrence probabilities and consequent losses of uncertain events are quantified. Extreme event samples were selected by means of conditional expectation to obtain the quantified description of each risk. Three ways of analyzing the quantified results are proposed.
{"title":"A Study of Risk Analysis Methods Based on Uncertain Event Analysis in Strategy Decision-Making","authors":"Rui Chu, Dongdong Yan, YouFei Cai, ChiFei Zhou","doi":"10.1109/ICTAI.2011.147","DOIUrl":"https://doi.org/10.1109/ICTAI.2011.147","url":null,"abstract":"A kind of strategy decision-making risk analysis method is put forward based on uncertain event analysis, on the basis of potential uncertain events confronted in decision-makings. In this method, a model is constructed to generate the decision-making option space, with detailed generation ways based on the risks of uncertain events. By comparing the potential capacities of the enemy, self-army and the third party, the occurrence probabilities and consequent losses of uncertain events are quantified. Extreme event samples were selected by means of conditional expectation to obtain the quantified description of each risk. Three ways of analyzing the quantified results are proposed.","PeriodicalId":332661,"journal":{"name":"2011 IEEE 23rd International Conference on Tools with Artificial Intelligence","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121323074","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：