Showing papers in "Artificial Intelligence Review in 2011"

An optimizing BP neural network algorithm based on genetic algorithm

Abstract: A back-propagation (BP) neural network has good self-learning, self-adapting and generalization ability, but it may easily get stuck in a local minimum, and has a poor rate of convergence. Therefore, a method to optimize a BP algorithm based on a genetic algorithm (GA) is proposed to speed the training of BP, and to overcome BP's disadvantage of being easily stuck in a local minimum. The UCI data set is used here for experimental analysis and the experimental result shows that, compared with the BP algorithm and a method that only uses GA to learn the connection weights, our method that combines GA and BP to train the neural network works better; is less easily stuck in a local minimum; the trained network has a better generalization ability; and it has a good stabilization performance.

A review on particle swarm optimization algorithms and their applications to data clustering

Abstract: Data clustering is one of the most popular techniques in data mining. It is a method of grouping data into clusters, in which each cluster must have data of great similarity and high dissimilarity with other cluster data. The most popular clustering algorithm K-mean and other classical algorithms suffer from disadvantages of initial centroid selection, local optima, low convergence rate problem etc. Particle Swarm Optimization (PSO) is a population based globalized search algorithm that mimics the capability (cognitive and social behavior) of swarms. PSO produces better results in complicated and multi-peak problems. This paper presents a literature survey on the PSO application in data clustering. PSO variants are also described in this paper. An attempt is made to provide a guide for the researchers who are working in the area of PSO and data clustering.

The variants of the harmony search algorithm: an overview

Abstract: The harmony search (HS) algorithm is a relatively new population-based metaheuristic optimization algorithm. It imitates the music improvisation process where musicians improvise their instruments' pitch by searching for a perfect state of harmony. Since the emergence of this algorithm in 2001, it attracted many researchers from various fields especially those working on solving optimization problems. Consequently, this algorithm guided researchers to improve on its performance to be in line with the requirements of the applications being developed. These improvements primarily cover two aspects: (1) improvements in terms of parameters setting, and (2) improvements in terms of hybridizing HS components with other metaheuristic algorithms. This paper presents an overview of these aspects, with a goal of providing useful references to fundamental concepts accessible to the broad community of optimization practitioners.

From symbolic to sub-symbolic information in question classification

Abstract: Question Answering (QA) is undoubtedly a growing field of current research in Artificial Intelligence. Question classification, a QA subtask, aims to associate a category to each question, typically representing the semantic class of its answer. This step is of major importance in the QA process, since it is the basis of several key decisions. For instance, classification helps reducing the number of possible answer candidates, as only answers matching the question category should be taken into account. This paper presents and evaluates a rule-based question classifier that partially founds its performance in the detection of the question headword and in its mapping into the target category through the use of WordNet. Moreover, we use the rule-based classifier as a features' provider of a machine learning-based question classifier. A detailed analysis of the rule-base contribution is presented. Despite using a very compact feature space, state of the art results are obtained.

A review of clonal selection algorithm and its applications

Abstract: Recently, clonal selection theory in the immune system has received the attention of researchers and given them inspiration to create algorithms that evolve candidate solutions by means of selection, cloning, and mutation procedures. Moreover, diversity in the population is enabled by means of the receptor editing process. The Clonal Selection Algorithm (CSA) in its canonical form and its various versions are used to solve different types of problems and are reported to perform better compared with other heuristics (i.e., genetic algorithms, neural networks, etc.) in some cases, such as function optimization and pattern recognition. Although the studies related with CSA are increasingly popular, according to our best knowledge, there is no study summarizing the basic features of these algorithms, hybrid algorithms, and the application areas of these algorithms all in one paper. Therefore, this study aims to summarize the powerful characteristics and general review of CSA. In addition, CSA based hybrid algorithms are reviewed, and open research areas are discussed for further research.

Combining bagging, boosting, rotation forest and random subspace methods

Abstract: Bagging, boosting, rotation forest and random subspace methods are well known re-sampling ensemble methods that generate and combine a diversity of learners using the same learning algorithm for the base-classifiers. Boosting and rotation forest algorithms are considered stronger than bagging and random subspace methods on noise-free data. However, there are strong empirical indications that bagging and random subspace methods are much more robust than boosting and rotation forest in noisy settings. For this reason, in this work we built an ensemble of bagging, boosting, rotation forest and random subspace methods ensembles with 6 sub-classifiers in each one and then a voting methodology is used for the final prediction. We performed a comparison with simple bagging, boosting, rotation forest and random subspace methods ensembles with 25 sub-classifiers, as well as other well known combining methods, on standard benchmark datasets and the proposed technique had better accuracy in most cases.

Locally linear embedding: a survey

Abstract: As a classic method of nonlinear dimensional reduction, locally linear embedding (LLE) is more and more attractive to researchers due to its ability to deal with large amounts of high dimensional data and its non-iterative way of finding the embeddings. However, several problems in the LLE algorithm still remain open, such as its sensitivity to noise, inevitable ill-conditioned eigenproblems, the lack of how to deal with the novel data, etc. The existing extensions are comprehensively reviewed and discussed classifying into different categories in this paper. Their strategies, advantages/disadvantages and performances are elaborated. By generalizing different tactics in various extensions related to different stages of LLE and evaluating their performances, several promising directions for future research have been suggested.

A survey of grammatical inference methods for natural language learning

Abstract: The high complexity of natural language and the huge amount of human and temporal resources necessary for producing the grammars lead several researchers in the area of Natural Language Processing to investigate various solutions for automating grammar generation and updating processes. Many algorithms for Context-Free Grammar inference have been developed in the literature. This paper provides a survey of the methodologies for inferring context-free grammars from examples, developed by researchers in the last decade. After introducing some preliminary definitions and notations concerning learning and inductive inference, some of the most relevant existing grammatical inference methods for Natural Language are described and classified according to the kind of presentation (if text or informant) and the type of information (if supervised, unsupervised, or semi-supervised). Moreover, the state of the art of the strategies for evaluation and comparison of different grammar inference methods is presented. The goal of the paper is to provide a reader with introduction to major concepts and current approaches in Natural Language Learning research.

Feature selection for support vector machines with RBF kernel

Abstract: Linear kernel Support Vector Machine Recursive Feature Elimination (SVM-RFE) is known as an excellent feature selection algorithm. Nonlinear SVM is a black box classifier for which we do not know the mapping function $${\Phi}$$ explicitly. Thus, the weight vector w cannot be explicitly computed. In this paper, we proposed a feature selection algorithm utilizing Support Vector Machine with RBF kernel based on Recursive Feature Elimination(SVM-RBF-RFE), which expands nonlinear RBF kernel into its Maclaurin series, and then the weight vector w is computed from the series according to the contribution made to classification hyperplane by each feature. Using $${w_i^2}$$ as ranking criterion, SVM-RBF-RFE starts with all the features, and eliminates one feature with the least squared weight at each step until all the features are ranked. We use SVM and KNN classifiers to evaluate nested subsets of features selected by SVM-RBF-RFE. Experimental results based on 3 UCI and 3 microarray datasets show SVM-RBF-RFE generally performs better than information gain and SVM-RFE.

Semantic web reasoners and languages

Abstract: Semantic web reasoners and languages enable the semantic web to function. Some of the latest reasoning models developed in the last few years are: DLP, FaCT, RACER, Pellet, MSPASS, CEL, Cerebra Engine, QuOnto, KAON2, HermiT and others. Some software tools such as Protege, Jena and others also have been developed, which provide inferencing as well as ontology development and management environments. These reasoners usually differ in their inference procedures, supporting logic, completeness of reasoning, expressiveness and implementation languages. Various semantic web languages with increasing expressive power continue to be developed for describing web services. We survey the some of the more recent languages like OWL-S (Ontology Web Language-Schema), WSML (Web Service Modeling Language), SWRL (Semantic Web Rule Language) and others that have been tested in early use. We also survey semantic web reasoners and their relationship to these languages.

A survey: hybrid evolutionary algorithms for cluster analysis

Abstract: Clustering is a popular data analysis and data mining technique. It is the unsupervised classification of patterns into groups. Many algorithms for large data sets have been proposed in the literature using different techniques. However, conventional algorithms have some shortcomings such as slowness of the convergence, sensitive to initial value and preset classed in large scale data set etc. and they still require much investigation to improve performance and efficiency. Over the last decade, clustering with ant-based and swarm-based algorithms are emerging as an alternative to more traditional clustering techniques. Many complex optimization problems still exist, and it is often very difficult to obtain the desired result with one of these algorithms alone. Thus, robust and flexible techniques of optimization are needed to generate good results for clustering data. Some algorithms that imitate certain natural principles, known as evolutionary algorithms have been used in a wide variety of real-world applications. Recently, much research has been proposed using hybrid evolutionary algorithms to solve the clustering problem. This paper provides a survey of hybrid evolutionary algorithms for cluster analysis.

Evolutionary multi objective optimization for rule mining: a review

Abstract: Evolutionary multi objective optimization (EMOO) systems are evolutionary systems which are used for optimizing various measures of the evolving system. Rule mining has gained attention in the knowledge discovery literature. The problem of discovering rules with specific properties is treated as a multi objective optimization problem. The objectives to be optimized being the metrics like accuracy, comprehensibility, surprisingness, novelty to name a few. There are a variety of EMOO algorithms in the literature. The performance of these EMOO algorithms is influenced by various characteristics including evolutionary technique used, chromosome representation, parameters like population size, number of generations, crossover rate, mutation rate, stopping criteria, Reproduction operators used, objectives taken for optimization, the fitness function used, optimization strategy, the type of data, number of class attributes and the area of application. This study reviews EMOO systems taking the above criteria into consideration. There are other hybridization strategies like use of intelligent agents, fuzzification, meta data and meta heuristics, parallelization, interactiveness with the user, visualization, etc., which further enhance the performance and usability of the system. Genetic Algorithms (GAs) and Genetic Programming (GPs) are two widely used evolutionary strategies for rule knowledge discovery in Data mining. Thus the proposed study aims at studying the various characteristics of the EMOO systems taking into consideration the two evolutionary strategies of Genetic Algorithm and Genetic programming.

Context-aware recommendation using rough set model and collaborative filtering

Abstract: Context has been identified as an important factor in recommender systems. Lots of researches have been done for context-aware recommendation. However, in current approaches, the weights of contextual information are the same, which limits the accuracy of the results. This paper aims to propose a context-aware recommender system by extracting, measuring and incorporating significant contextual information in recommendation. The approach is based on rough set theory and collaborative filtering. It involves a three-steps process. At first, significant attributes to represent contextual information are extracted and measured to identify recommended items based on rough set theory. Then the users' similarity is measured in a target context consideration. Furthermore collaborative filtering is adopted to recommend appropriate items. The evaluation experiments show that the proposed approach is helpful to improve the recommendation quality.

A survey of keyword spotting techniques for printed document images

Abstract: This paper attempts to provide a survey of the past researches on character based as keyword based approaches used for retrieving information from document images. This survey also provides insights into the strengths and weaknesses of current techniques, relevancy lies between each technique and also the guidance in choosing the area that future work on document image retrieval could address.

Data mining applications in hydrocarbon exploration

Abstract: This paper presents a review of the use of intelligent data analysis techniques in Hydrocarbon Exploration. The term "intelligent" is used in its broadest sense. The process of hydrocarbon exploration exploits data which have been collected from different sources. Different dimensions of data are analyzed by using Statistical Analysis, Data Mining, Artificial Neural Networks and Artificial Intelligence. This review is meant not only to describe the evolution of intelligent data analysis techniques used in different phases of hydrocarbon exploration but also signifying the growing use of Data Mining in various application domains; we avoided a general review of Data Mining and other intelligent data analysis techniques in this paper. The volume of general literature might affect the precision of our view regarding the application of these techniques in hydrocarbon exploration. The review reveals the suitability of existing techniques to data collected from diverse sources in addition to the use of analytical techniques for the process of hydrocarbon exploration.

A review: accuracy optimization in clustering ensembles using genetic algorithms

Abstract: The clustering ensemble has emerged as a prominent method for improving robustness, stability, and accuracy of unsupervised classification solutions. It combines multiple partitions generated by different clustering algorithms into a single clustering solution. Genetic algorithms are known as methods with high ability to solve optimization problems including clustering. To date, significant progress has been contributed to find consensus clustering that will yield better results than existing clustering. This paper presents a survey of genetic algorithms designed for clustering ensembles. It begins with the introduction of clustering ensembles and clustering ensemble algorithms. Subsequently, this paper describes a number of suggested genetic-guided clustering ensemble algorithms, in particular the genotypes, fitness functions, and genetic operations. Next, clustering accuracies among the genetic-guided clustering ensemble algorithms is compared. This paper concludes that using genetic algorithms in clustering ensemble improves the clustering accuracy and addresses open questions subject to future research.

Architectural space planning using evolutionary computing approaches: a review

Abstract: This paper presents various applications of evolutionary computing approach for architectural space planning problem. As such the problem of architectural space planning is NP-complete. Finding an optimal solution within a reasonable amount of time for these problems is impossible. However for architectural space planning problem we may not be even looking for an optimal but some feasible solution based on varied parameters. Many different computing approaches for space planning like procedural algorithms, heuristic search based methods, genetic algorithms, fuzzy logic, and artificial neural networks etc. have been developed and are being employed. In recent years evolutionary computation approaches have been applied to a wide variety of applications as it has the advantage of giving reasonably acceptable solution in a reasonable amount of time. There are also hybrid systems such as neural network and fuzzy logic which incorporates the features of evolutionary computing paradigm. The present paper aims to compare the various aspects and merits/demerits of each of these methods developed so far. Sixteen papers have been reviewed and compared on various parameters such as input features, output produced, set of constraints, scope of space coverage-single floor, multi-floor and urban spaces. Recent publications emphasized on energy aspect as well. The paper will help the better understanding of the Evolutionary computing perspective of solving architectural space planning problem. The findings of this paper provide useful insight into current developments and are beneficial for those who look for automating architectural space planning task within given design constraints.

A discrete shuffled frog optimization algorithm

Abstract: The shuffled frog leaping (SFL) optimization algorithm has been successful in solving a wide range of real-valued optimization problems. In this paper we present a discrete version of this algorithm and compare its performance with a SFL algorithm, a binary genetic algorithm (BGA), and a discrete particle swarm optimization (DPSO) algorithm on seven low dimensional and five high dimensional benchmark problems. The obtained results demonstrate that our proposed algorithm, i.e. the DSFL, outperforms the BGA and the DPSO in terms of both success rate and speed. On low dimensional functions and for large values of tolerance the DSFL is slower than the SFL, but their success rates are equal. Part of this slowness could be attributed to the extra bits used for data coding. By increasing number of variables and the required precision of answer, the DSFL performs very well in terms of both speed and success rate. For high dimensional problems, for intrinsically discrete problems, also when the required precision of answer is high, the DSFL is the most efficient method.

BSPNN: boosted subspace probabilistic neural network for email security

Abstract: In the modern age of Internet connectivity, advanced information systems have accumulated huge volumes of data. Such fast growing, tremendous amount of data, collected and stored in large databases has far exceeded our human ability to comprehend without proper tools. There has been a great deal of research conducted to explore the potential applications of Machine Learning technologies in Security Informatics. This article studies the Network Security Detection problems in which predictive models are constructed to detect network security breaches such as spamming. Due to overwhelming volume of data, complexity and dynamics of computer networks and evolving cyber threats, current security systems suffer limited performance with low detection accuracy and high number of false alarms. To address such performance issues, a novel Machine Learning algorithm, namely Boosted Subspace Probabilistic Neural Network (BSPNN), has been proposed which combines a Radial Basis Function Neural Network with an innovative diversity-based ensemble learning framework. Extensive empirical analyses suggested that BSPNN achieved high detection accuracy with relatively small computational complexity compared with other conventional detection methods.

Document image analysis: issues, comparison of methods and remaining problems

Abstract: Image analysis is an interesting research area with a large variety of challenging applications. Researchers have worked from decades on this topic, as witnessed by the scientific literature. However, document image analysis is the special case in image analysis as their spatial properties are different from natural images. Therefore, the main focus of this paper is to describe image denoising issues in general and document image issues in particular. Since the field of document processing is relatively new, it is also dynamic, so current methods have room for improvement and innovations are still being made. Several algorithms proposed in the literature are described. Critical discussions are reported about the current status of the field and open problems are highlighted. It is also demonstrated that, there are rarely definitive techniques for all cases of a certain problem. We surveyed the state of art, analyzed recent trends and tried to identify challenges for future research in this field.

Mining data streams with concept drifts using genetic algorithm

Abstract: Recent research shows that rule based models perform well while classifying large data sets such as data streams with concept drifts. A genetic algorithm is a strong rule based classification algorithm which is used only for mining static small data sets. If the genetic algorithm can be made scalable and adaptable by reducing its I/O intensity, it will become an efficient and effective tool for mining large data sets like data streams. In this paper a scalable and adaptable online genetic algorithm is proposed to mine classification rules for the data streams with concept drifts. Since the data streams are generated continuously in a rapid rate, the proposed method does not use a fixed static data set for fitness calculation. Instead, it extracts a small snapshot of the training example from the current part of data stream whenever data is required for the fitness calculation. The proposed method also builds rules for all the classes separately in a parallel independent iterative manner. This makes the proposed method scalable to the data streams and also adaptable to the concept drifts that occur in the data stream in a fast and more natural way without storing the whole stream or a part of the stream in a compressed form as done by the other rule based algorithms. The results of the proposed method are comparable with the other standard methods which are used for mining the data streams.

An incremental ensemble of classifiers

Abstract: Along with the increase of data and information, incremental learning ability turns out to be more and more important for machine learning approaches. The online algorithms try not to remember irrelevant information instead of synthesizing all available information (as opposed to classic batch learning algorithms). Today, combining classifiers is proposed as a new road for the improvement of the classification accuracy. However, most ensemble algorithms operate in batch mode. For this reason, we propose an incremental ensemble that combines five classifiers that can operate incrementally: the Naive Bayes, the Averaged One-Dependence Estimators (AODE), the 3-Nearest Neighbors, the Non-Nested Generalised Exemplars (NNGE) and the Kstar algorithms using the voting methodology. We performed a large-scale comparison of the proposed ensemble with other state-of-the-art algorithms on several datasets and the proposed method produce better accuracy in most cases.

Improving car environmental and operational characteristics using a multifunctional fuel additive

Abstract: Vehicle modernization has been developed towards the growing necessities of speed, power, efficiency, ergonomics, and design etc. The requirements, nowadays of environmental safety and operational efficiency of vehicles are being brought to the forefront. The aim of this work is to increase efficiency and reduce the harmful environmental impact of automobile transport by improving the quality of fuels during its operation. The improvement of the quality of fuels by means of highly-effective additives is the most rapidly implemented and lowcost method. According to the settled requirements of the properties of additives and the analysis of the catalytic and physicochemical properties of the substances, the universal content of the additive to gasoline and diesel fuel was found and the technology of its production was proposed. In addition, the additive was thoroughly tested in the laboratory, test bench, traffic operation and experimental-industrial checkout. It was found that the application of the additive in minor quantities significantly improves operational and environmental properties of fuels and engine characteristics.

Multi-agent based simulations using fast multipole method: application to large scale simulations of flocking dynamical systems

Abstract: This article introduces a novel approach to increase the performances of multi-agent based simulations. We focus on a particular kind of multi-agent based simulation where a collection of interacting autonomous situated entities evolve in a situated environment. Our approach combines the fast multipole method coming from computational physics with agent-based microscopic simulations. The aim is to speed up the execution of a multi-agent based simulation while controlling the precision of the associated approximation. This approach may be considered as the first step of a larger effort aiming at designing a generic kernel to support efficient large-scale multi-agent based simulations. This approach is illustrated in this paper by the simulation of large scale flocking dynamical systems.

Particulates in the atmosphere of Makkah and Mina valley during the Ramadan and Hajj seasons of 2004 and 2005

Abstract: This work has been devoted to study TSP, PM10 and PM2.5 in the atmosphere of Makkah and the Mina valley during the Ramadan and Hajj periods, 1424 and 1425 H. On the occasion of Hajj, about 2.5 million persons gather in Makkah and move to Mina valley (4 km 2 ), 7 km outside east of Makkah. Pilgrims spend 3 nights in the valley. Congested traffic and the high rates of emissions in such a valley of small area coupled with severe weather conditions, make the area ideal for the accumulation of air pollutants. The present investigation shows that the diurnal cycle of PM10 in air coincides with the pattern of traffic movements. Particulate matters (PM10) daily concentrations in the atmosphere of the Mina valley ranged between 191–262µg/m 3 during the presence of the pilgrims in Mina compared to the European standard of 50µg/m 3 . These concentrations represent 34%–40% of TSP. These high PM10 concentrations are due to the massive transportation movements at Mina valley. Moreover, TSP concentrations reached 665µg/m 3 in the Makkah atmosphere during the last ten days of Ramadan compared to the Saudi standard of 340µg/m

What agents can do in workflow management systems

Abstract: Workflow management systems are an emerging category of information systems, currently under dynamic evolution. On the other hand software agents are a distinct research area as well as an emerging paradigm for information systems design and development. This paper tries to examine the integration of these two fields; reveal the stimulation and the advantages of such a mixing. Popular standards of the workflow management field are used to derive a classification scheme, which is exploited to map existing approaches. As a significant number of existing approaches is reviewed, a plethora of integration patterns are identified and grouped according to the proposed classification scheme. The overall goal of the paper is to clear the vague picture of the consolidation of workflow management systems and software agents and to provide an unifying framework for this intersected area.

Non-parametric Nature Of Ground-level OzoneAnd Its Dependence On Nitrogen Oxides (NOx):A View Point Of Vehicular Emissions

Abstract: Ground-level ozone has been studied extensively using classic parametric statistics (most commonly conventional linear regression). Very few researchers have considered ozone distributions and even those that do tend to apply parametric techniques. This study assesses ground-level ozone distributions at six locations in the UK and characterises the correlation of nitrogen oxides (NOx) and ozone at a roadside location. The distribution of ozone is investigated, applying Shapiro-Wilk test and graphical presentations. The histograms are right skewed and show maximum frequency at ozone mixing ratios from 0 to 5 ppb (particularly at urban centers and roadsides locations), which is probably caused by high levels of freshly produced NOx associated with road traffic. There is evidence that ground level ozone is not normally distributed (p-values < 0.05). NOx is a dominant sink for ozone at urban and roadside sites due to its ozone scavenging effects. Consistent with literature ozone is negatively correlated with NOx. The negative correlation is stronger at low NOx levels (up to approximately 80 ppb 24 hour mean, Spearman correlation coefficient R is ‘-0.72’) and becomes weaker as NOx levels increase (over 80 ppb R value is ‘-0.53’). When NOx mixing ratios reach approximately 200 ppb or over the correlations become positive. This study investigates how the associations of ozone and NOx vary at different levels of their mixing ratios and suggests that due to the non-normal distribution of ozone, nonparametric statistics should be applied for ozone modelling.

Discrete bee algorithms and their application in multivariable function optimization

Abstract: In this paper we present four discrete versions of two different existing honey bee optimization algorithms: the discrete artificial bee colony algorithm (DABC) and three versions of the discrete fast marriage in honey bee optimization algorithm (DFMBO1, DFMBO2, and DFMBO3). In these discretized algorithms we have utilized three logical operators, i.e. OR, AND and XOR operators. Then we have compared performances of our algorithms and those of three other bee algorithms, i.e. the artificial bee colony (ABC), the queen bee (QB), and the fast marriage in honey bee optimization (FMBO) on four benchmark functions for various numbers of variables up to 100. The obtained results show that our discrete algorithms are faster than other algorithms. In general, when precision of answer and number of variables are low, the difference between our new algorithms and the other three algorithms is small in terms of speed, but by increasing precision of answer and number of variables, the needed number of function evaluations for other algorithms increases beyond manageable amounts, hence their success rates decrease. Among our proposed discrete algorithms, the DFMBO3 is always fast, and achieves a success rate of 100% on all benchmarks with an average number of function evaluations not more than 1010.

Impact of urban planning alternatives on air quality: URBAIR model application

Abstract: In the last decades, the study of the urban structure impacts on the quality of life and on the environment became a key issue for urban sustainability. Nowadays the relevance of urban planning for the improvement of the interactions between different land uses and economic activities, and also towards a more sustainable urban metabolism, is consensually accepted. A major interest relies on understanding the role of planning on induced mobility patterns and thereafter on air quality, particularly related with the increasing use of private cars. This is one of the main objectives of BRIDGE, a research project funding by the European Commission under the 7 th Framework Programme and focused on “SustainaBle uRban plannIng Decision support accountinG for urban mEtabolism”. In this scope, and to evaluate the impact on air quality due to different city planning alternatives (PA), the urban scale air quality modelling system URBAIR was applied to selected areas in Helsinki (Finland), Athens (Greece) and Gliwice (Poland), to estimate traffic related emissions and induced pollutant concentration of different air pollutants, in a hourly basis for the entire year of 2008. For the Helsinki study case the results suggest that urban traffic and building placement considered on the different PA have an influence on local air quality despite no significant concentration levels. In the Athens case study some PA induce a decrease on traffic flows with an improvement of the air quality over the domain. On the contrary, other leads to an increase of PM10 in selected hotspots. The simulations for the Gliwice study case show minor changes between the baseline and the PA, since the proposed interventions do not imply major changes in traffic flows.

State of the art versus classical clustering for unsupervised word sense disambiguation

Abstract: This paper ultimately discusses the importance of the clustering method used in unsupervised word sense disambiguation. It illustrates the fact that a powerful clustering technique can make up for lack of external knowledge of all types. It argues that feature selection does not always improve disambiguation results, especially when using an advanced, state of the art method, hereby exemplified by spectral clustering. Disambiguation results obtained when using spectral clustering in the case of the main parts of speech (nouns, adjectives, verbs) are compared to those of the classical clustering method given by the Naive Bayes model. In the case of unsupervised word sense disambiguation with an underlying Naive Bayes model feature selection performed in two completely different ways is surveyed. The type of feature selection providing the best results (WordNet-based feature selection) is equally being used in the case of spectral clustering. The conclusion is that spectral clustering without feature selection (but using its own feature weighting) produces superior disambiguation results in the case of all parts of speech.