Automated negotiation between artificial agents is essential to deploy Cognitive Computing and Internet of Things. The behavior of a negotiation agent depends significantly on the influence of environmental conditions or contextual variables, since they affect not only a given agent preferences and strategies, but also those of other agents. Despite this, the existing literature on automated negotiation is scarce about how to properly account for the effect of context-relevant variables in learning and evolving strategies. In this paper, a novel context-driven representation for automated negotiation is introduced. Also, a simple negotiation agent that queries available information from its environment, internally models contextual variables, and learns how to take advantage of this knowledge by playing against himself using reinforcement learning is proposed. Through a set of episodes against other negotiation agents in the existing literature, it is shown using our context-aware agent that it makes no sense to negotiate without taking context-relevant variables into account. Our context-aware negotiation agent has been implemented in the GENIUS environment, and results obtained are significant and quite revealing.

Considerable improvements in the technology and performance of SAT solvers has made their use possible for the resolution of various problems in artificial intelligence, and among them that of generating plans. Recently, promising Quantified Boolean Formula (QBF) solvers have been developed and we may expect that in a near future they become as efficient as SAT solvers. So, it is interesting to use QBF language that allows us to produce more compact encodings. We present in this article a translation from STRIPS planning problems into quantified propositional formulas. We introduce two new Compact Tree Encodings: CTE-EFA based on Explanatory frame axioms, and CTE-OPEN based on causal links. Then we compare both of them to CTE-NOOP based on No-op Actions proposed in [Cashmore et al. 2012]. In terms of execution time over benchmark problems, CTE-EFA and CTE-OPEN always performed better than CTE-NOOP.

Un problema que enfrentan los organismos operadores de agua, es el cumplimiento de la normatividaden la calidad del agua residual tratada. Por lo que es recomendable implementar estrategias que favorezcan elcumplimiento de las regulaciones. La minería de datos es una herramienta que permite predecir la calidad del aguaen el efluente de los sistemas de tratamiento. En el presente estudio se propone un criterio para el pre procesado dedatos donde se consideraron variables nominales. Luego se aplicó el sistema de minería de datos (clasificación)para definir la predicción de la calidad del agua. Se aplicaron los siguientes clasificadores: OneR; decisión tables,J48, árbol de decisión de un solo nivel; PART y LMT. Los resultados muestran que el mejor algoritmo fue el J48:87.35 % de instancias correctamente clasificadas. El árbol de decisión determinó dos reglas para el cumplimientocon la normatividad. Es importante indicar que a la fecha existen procedimientos con minería de datos parapredecir la calidad del efluente de un sistema de tratamiento, pero utilizan estrictamente variables numéricas;mientras que en el presente trabajo se utilizaron variables nominales, finalmente se discuten los resultados y seindican los procesos industriales que generan materia orgánica y otros contaminantes.

The analysis of group dynamics is extremely useful for understanding and predicting the performance of teamwork’s, since in this context, collaboration problems can naturally arise. Artificial intelligence, and specially machine learning techniques, enables automating the observation process and the analysis of groups of users who use an online collaborative platform. Among the online collaborative platforms available, games are an attractive alternative for all audiences that enable capturing the players’ behavior by observing their social interactions, while engaging them in a pleasant activity. In this paper, we present experimental results of classifying observed conversations in an online game to collaborative behaviors, guided by the Interaction Process Analysis, a theory for categorizing social interactions. The proposed automation of the classification process can be used to assist teachers or team leaders to detect alterations in the balance of group reactions and to improve their performance by indicating actions to improve the balance.


 
 
 Nowadays, progress is constant and inherent to a living society. This may occur in different arenas, namely in mathematical evaluation and healthcare. Assistive technologies are a topic under this evolution, being extremely important in helping users with diminished capabilities (physical, sensory, intellectual). These technologies assist people in tasks that were difficult or impossible to execute. A common diminished task is orientation, which is crucial for the user autonomy. The adaptation to such technologies should require the minimum effort possible in order to enable the person to use devices that convey assistive functionalities. There are several solutions that help a human being to travel between two different locations, however their authors are essentially concerned with the guidance method, giving special attention to the user interface. The CogHelper system aims to overcome these systems by applying a framework of Speculative Computation, which adds a prediction feature for the next user movement giving an anticipation ability to the system. Thus, an alert is triggered before the user turn towards an incorrect path. The travelling path is also adjusted to the user preferences through a trajectory mining module.
 
 

Nowadays, making strategic decisions in a sensitive sector such as energy planning that usually requiresallocating huge funds, time, and resources is a difficult task. For instance, prioritizing a set of Renewable EnergySources (RES) is a complex multi-dimensional task that typically involves a range of conflicting criteria featuringdifferent forms of evaluation data in an uncertain decision-making environment. This process is aligned withseveral sources that can be uncertain, including imprecise information, limited domain knowledge from decisionmakers,and failures to provide accurate judgments from experts. In this study, we propose to use the EvidentialReasoning (ER) approach to manage the expanding complexities and uncertainties in RES prioritization problem.The ER approach is employed as a multiple criteria framework to assess the appropriateness regarding the use ofdifferent renewable energy technologies. A case study is provided to illustrate the implementation process. Resultsshow that using the ER approach when assessing the sustainability of different RES under uncertainty allowsproviding robust decisions, which brings out a more accurate, effective, and better-informed decision-making toolto conduct the evaluation process.

Reasoning with preference information is a common human activity. As modelling human reasoning is one of the main objectives of AI, reasoning with preferences is an important topic in various fields of AI, such as Knowledge Representation and Reasoning (KR). Argumentation is one particular branch of KR that concerns, among other tasks, modelling common-sense reasoning with preferences. A key issue there, is the lack of consensus on how to deal with preferences. Witnessing this is a multitude of proposals on how to formalise reasoning with preferences in argumentative terms. As a commonality, however, formalisms of argumentation with preferences tend to fulfil various criteria of 'rational' reasoning, notwithstanding the fact that human reasoning is often not 'rational', yet seemingly 'intuitive'. In this paper, we study how several formalisms of argumentation with preferences model human intuition behind a particular common-sense reasoning problem. More specifically, we present a common-sense scenario of reasoning with rules and preferences, complemented with a survey of decisions made by human respondents that indicates an 'intuitive' solution, and analyse how this problem is tackled in argumentation. We conclude that most approaches to argumentation with preferences afford a 'rational' solution to the problem, and discuss one recent formalism that yields the 'intuitive' solution instead. We argue that our results call for advancements in the area of argumentation with preferences in particular, as well as for further studies of reasoning with preferences in AI at large.

This paper describes how a forward model could be applied in a manipulator robot to accomplish the task of following a moving target. The forward model has been implemented in the puma 560 robot manipulator in simulation after a babbling motor phase using ANFIS neural networks. The forward model delivers a rough estimation of the position in the operational space of a moving target. Using this information a Cartesian controller tracks the moving target. An implementation of the proposed architecture and the Piepmeir algorithm for the problem of following a moving target is also shown in the paper. The control architecture proposed in this paper was also tested with MLP and RBF neural networks. Results and simulations are shown to demonstrate the applicability of our proposed architecture for tracking a moving target.

The hotel exploitation, while continuing to satisfy the customers, needs to decrease the requests of electric power as the principal energy carrier. Solving issues regarding the occupation of a hotel integrally, taking the air conditioning as center of attention, which demands the bigger consumptions of electricity, results in a complex task. To solve this issue, a procedure was implemented to optimize the operation of the water-chilled systems. The procedure integrates an energy model with a strategy of low occupation following energetic criteria based on combinatorial-evolutionary criteria. To classify the information, the formulation of the tasks and the synthesis of the solutions, a methodology of analysis and synthesis of engineering is used. The energetic model considers the variability of the local climatology and the occupation of the selected rooms, and includes: the thermal model of the building obtained by means of artificial neural networks, the hydraulic model and the model of the compression work. These elements allow to find the variable of decision occupation, performing intermediate calculations to obtain the velocity of rotation in the centrifugal pump and the output temperature of the cooler water, minimizing the requirements of electric power in the water-chilled systems. To evaluate the states of the system, a combinatorial optimization is used through the following methods: simple exhaustive, stepped exhaustive or genetic algorithm depending on the quantity of variants of occupation. All calculation tasks and algorithms of the procedure were automated through a computer application.