Vector-space word representations obtained from neural network models have been shown to enable s... more Vector-space word representations obtained from neural network models have been shown to enable semantic operations based on vector arithmetic. In this paper, we explore the existence of similar information on vector representations of images. For that purpose we define a methodology to obtain large, sparse vector representations of image classes, and generate vectors through the state-of-the-art deep learning architecture GoogLeNet for 20K images obtained from ImageNet. We first evaluate the resultant vector-space semantics through its correlation with Word-Net distances, and find vector distances to be strongly correlated with linguistic semantics. We then explore the location of images within the vector space, finding elements close in WordNet to be clustered together, regardless of significant visual variances (e.g., 118 dog types). More surprisingly , we find that the space unsupervisedly separates complex classes without prior knowledge (e.g., living things). Afterwards, we consider vector arithmetics. Although we are unable to obtain meaningful results on this regard, we discuss the various problem we encountered, and how we consider to solve them. Finally, we discuss the impact of our research for cognitive systems, focusing on the role of the architecture being used.
ABSTRACT Cities concentrate enough Social Network (SN) activity to empower rich models. We presen... more ABSTRACT Cities concentrate enough Social Network (SN) activity to empower rich models. We present an approach to event discovery based on the information provided by three SN, minimizing the data properties used to maximize the total amount of usable data. We build a model of the normal city behavior which we use to detect abnor- mal situations (events). After collecting half a year of data we show examples of the events detected and introduce some applications.
ABSTRACT Cities concentrate enough Social Network (SN) activity to empower rich models. We presen... more ABSTRACT Cities concentrate enough Social Network (SN) activity to empower rich models. We present an approach to event discovery based on the information provided by three SN, minimizing the data properties used to maximize the total amount of usable data. We build a model of the normal city behavior which we use to detect abnor- mal situations (events). After collecting half a year of data we show examples of the events detected and introduce some applications.
International Journal on Artificial Intelligence Tools, 2014
ABSTRACT Nowadays Web services are one of the most commonly used technologies for distributed sys... more ABSTRACT Nowadays Web services are one of the most commonly used technologies for distributed system integration and interoperability. Recently, a new communication technique called WebSockets has appeared for implementing Web services that offer a true server-push mechanism. In this paper we present a system based on it to manage agent communications, and we motivate its use in an Assistive Technologies use case. A new WebSocket Message Transport Protocol is described and an implementation for the Jade platform is presented. Subscription based agent interaction for the use case is presented, along with a comparison between WebSocket and HTTP Message Transport Protocols.
Water quality management policies on a river basin scale are of special importance in order to pr... more Water quality management policies on a river basin scale are of special importance in order to prevent and/or reduce environmental pollution caused by human sources. Industrial effluents are a priority issue particularly in Urban Wastewater Systems (UWS) that receive mixed household and industrial wastewaters, apart from rainfall water. In this paper, we present an analysis and implementation of normative agents that capture concrete regulations of the Catalan pollution-prevention policies. The implementation of the ...
Journal of Experimental & Theoretical Artificial Intelligence, 2015
ABSTRACT Location Based Social Networks (LBSN) have become an interesting source for mining user ... more ABSTRACT Location Based Social Networks (LBSN) have become an interesting source for mining user behavior. These networks (e.g. Twitter, Instagram or Foursquare) collect spatio-temporal data from users in a way that they can be seen as a set of collective and distributed sensors on a geographical area. Processing this information in different ways could result in patterns useful for several application domains. These patterns include simple or complex user visits to places in a city or groups of users that can be described by a common behavior. The domains of application range from the recommendation of points of interest to visit and route planning for touristic recommender systems to city analysis and planning. This paper presents the analysis of data collected for several months from such LBSN inside the geographical area of two large cities. The goal is to obtain by means of unsupervised data mining methods sets of patterns that describe groups of users in terms of routes, mobility patterns and behavior profiles that can be useful for city analysis and mobility decisions.
The use of Horizontal Subsurface Constructed Wetlands (HSCWs) for treating wastewaters in small c... more The use of Horizontal Subsurface Constructed Wetlands (HSCWs) for treating wastewaters in small communities has increased in the last years due to HSCW's ecological singularities. Unfortunately, the same singularities that differentiate HSCWs complicate any attempt to develop models and produce generic decision-support systems for them. Classical mathematical and statistical approaches used in other Wastewater Treatment Plants do not properly fit the particularities of HSCW and provide little insight in the domain of HSCW. ...
Water quality management policies on a river basin scale are of special importance in order to pr... more Water quality management policies on a river basin scale are of special importance in order to prevent and/or reduce pollution of several human sources into the environment. Industrial effluents represent a priority issue particularly in Urban Wastewater Systems (UWS) that receive mixed household and industrial wastewaters, apart from rainfall water. In this paper, we present an analysis and an implementation of normative agents which capture concrete regulations of the Catalan pollution-prevention policies. The implementation of the normative agents is based on situation calculus.
Assistive Technologies (AT) are an application area where several Artificial Intelligence techniq... more Assistive Technologies (AT) are an application area where several Artificial Intelligence techniques and tools have been successfully applied to support elderly or impeded people on their daily activities. However, approaches to AT tend to center in the user-tool interaction, neglecting the user's connection with its social environment (such as caretakers, relatives and health professionals) and the possibility to monitor undesired behaviour providing both adaptation to a dynamic environment and early response to potentially dangerous situations. In previous work we have presented COAALAS, an intelligent social and norm-aware device for elderly people that is able to autonomously organize, reorganize and interact with the different actors involved in elderly-care, either human actors or other devices. In this paper we put our work into context, by first examining what are the desirable properties of such a system, analysing the state-of-the-art on the relevant topics, and verifying the validity of our proposal in a larger context that we call AVICENA. AVICENA's aim is develop a semi-autonomous (collaborative) tool to promote monitored, intensive, extended and personalized therapeutic regime adherence at home based on adaptation techniques.
The use of Horizontal Subsurface Constructed Wetlands (HSCWs) for treating wastewaters in small c... more The use of Horizontal Subsurface Constructed Wetlands (HSCWs) for treating wastewaters in small communities has increased in the last years due to HSCW's ecological singularities. Unfortunately, the same singularities that differentiate HSCWs complicate any attempt to develop models and produce generic decision-support systems for them. Classical mathematical and statistical approaches used in other Wastewater Treatment Plants do not properly fit the particularities of HSCW and provide little insight in the domain of HSCW. ...
English: The goal of this thesis is to study the viability of an IDSS based on a LFAC. At the end... more English: The goal of this thesis is to study the viability of an IDSS based on a LFAC. At the end, we intend to develop a complete IDSS application based on a LFAC which implements a real domain and which can be put to use in real scenarios. This thesis will focus on the problems combining two disciplines as di erent as IDSS and LFAC in order to make them work autonomously and transparently. We will work with the domain of a HSCW, a very common
Vector-space word representations obtained from neural network models have been shown to enable s... more Vector-space word representations obtained from neural network models have been shown to enable semantic operations based on vector arithmetic. In this paper, we explore the existence of similar information on vector representations of images. For that purpose we define a methodology to obtain large, sparse vector representations of image classes, and generate vectors through the state-of-the-art deep learning architecture GoogLeNet for 20K images obtained from ImageNet. We first evaluate the resultant vector-space semantics through its correlation with Word-Net distances, and find vector distances to be strongly correlated with linguistic semantics. We then explore the location of images within the vector space, finding elements close in WordNet to be clustered together, regardless of significant visual variances (e.g., 118 dog types). More surprisingly , we find that the space unsupervisedly separates complex classes without prior knowledge (e.g., living things). Afterwards, we consider vector arithmetics. Although we are unable to obtain meaningful results on this regard, we discuss the various problem we encountered, and how we consider to solve them. Finally, we discuss the impact of our research for cognitive systems, focusing on the role of the architecture being used.
ABSTRACT Cities concentrate enough Social Network (SN) activity to empower rich models. We presen... more ABSTRACT Cities concentrate enough Social Network (SN) activity to empower rich models. We present an approach to event discovery based on the information provided by three SN, minimizing the data properties used to maximize the total amount of usable data. We build a model of the normal city behavior which we use to detect abnor- mal situations (events). After collecting half a year of data we show examples of the events detected and introduce some applications.
ABSTRACT Cities concentrate enough Social Network (SN) activity to empower rich models. We presen... more ABSTRACT Cities concentrate enough Social Network (SN) activity to empower rich models. We present an approach to event discovery based on the information provided by three SN, minimizing the data properties used to maximize the total amount of usable data. We build a model of the normal city behavior which we use to detect abnor- mal situations (events). After collecting half a year of data we show examples of the events detected and introduce some applications.
International Journal on Artificial Intelligence Tools, 2014
ABSTRACT Nowadays Web services are one of the most commonly used technologies for distributed sys... more ABSTRACT Nowadays Web services are one of the most commonly used technologies for distributed system integration and interoperability. Recently, a new communication technique called WebSockets has appeared for implementing Web services that offer a true server-push mechanism. In this paper we present a system based on it to manage agent communications, and we motivate its use in an Assistive Technologies use case. A new WebSocket Message Transport Protocol is described and an implementation for the Jade platform is presented. Subscription based agent interaction for the use case is presented, along with a comparison between WebSocket and HTTP Message Transport Protocols.
Water quality management policies on a river basin scale are of special importance in order to pr... more Water quality management policies on a river basin scale are of special importance in order to prevent and/or reduce environmental pollution caused by human sources. Industrial effluents are a priority issue particularly in Urban Wastewater Systems (UWS) that receive mixed household and industrial wastewaters, apart from rainfall water. In this paper, we present an analysis and implementation of normative agents that capture concrete regulations of the Catalan pollution-prevention policies. The implementation of the ...
Journal of Experimental & Theoretical Artificial Intelligence, 2015
ABSTRACT Location Based Social Networks (LBSN) have become an interesting source for mining user ... more ABSTRACT Location Based Social Networks (LBSN) have become an interesting source for mining user behavior. These networks (e.g. Twitter, Instagram or Foursquare) collect spatio-temporal data from users in a way that they can be seen as a set of collective and distributed sensors on a geographical area. Processing this information in different ways could result in patterns useful for several application domains. These patterns include simple or complex user visits to places in a city or groups of users that can be described by a common behavior. The domains of application range from the recommendation of points of interest to visit and route planning for touristic recommender systems to city analysis and planning. This paper presents the analysis of data collected for several months from such LBSN inside the geographical area of two large cities. The goal is to obtain by means of unsupervised data mining methods sets of patterns that describe groups of users in terms of routes, mobility patterns and behavior profiles that can be useful for city analysis and mobility decisions.
The use of Horizontal Subsurface Constructed Wetlands (HSCWs) for treating wastewaters in small c... more The use of Horizontal Subsurface Constructed Wetlands (HSCWs) for treating wastewaters in small communities has increased in the last years due to HSCW's ecological singularities. Unfortunately, the same singularities that differentiate HSCWs complicate any attempt to develop models and produce generic decision-support systems for them. Classical mathematical and statistical approaches used in other Wastewater Treatment Plants do not properly fit the particularities of HSCW and provide little insight in the domain of HSCW. ...
Water quality management policies on a river basin scale are of special importance in order to pr... more Water quality management policies on a river basin scale are of special importance in order to prevent and/or reduce pollution of several human sources into the environment. Industrial effluents represent a priority issue particularly in Urban Wastewater Systems (UWS) that receive mixed household and industrial wastewaters, apart from rainfall water. In this paper, we present an analysis and an implementation of normative agents which capture concrete regulations of the Catalan pollution-prevention policies. The implementation of the normative agents is based on situation calculus.
Assistive Technologies (AT) are an application area where several Artificial Intelligence techniq... more Assistive Technologies (AT) are an application area where several Artificial Intelligence techniques and tools have been successfully applied to support elderly or impeded people on their daily activities. However, approaches to AT tend to center in the user-tool interaction, neglecting the user's connection with its social environment (such as caretakers, relatives and health professionals) and the possibility to monitor undesired behaviour providing both adaptation to a dynamic environment and early response to potentially dangerous situations. In previous work we have presented COAALAS, an intelligent social and norm-aware device for elderly people that is able to autonomously organize, reorganize and interact with the different actors involved in elderly-care, either human actors or other devices. In this paper we put our work into context, by first examining what are the desirable properties of such a system, analysing the state-of-the-art on the relevant topics, and verifying the validity of our proposal in a larger context that we call AVICENA. AVICENA's aim is develop a semi-autonomous (collaborative) tool to promote monitored, intensive, extended and personalized therapeutic regime adherence at home based on adaptation techniques.
The use of Horizontal Subsurface Constructed Wetlands (HSCWs) for treating wastewaters in small c... more The use of Horizontal Subsurface Constructed Wetlands (HSCWs) for treating wastewaters in small communities has increased in the last years due to HSCW's ecological singularities. Unfortunately, the same singularities that differentiate HSCWs complicate any attempt to develop models and produce generic decision-support systems for them. Classical mathematical and statistical approaches used in other Wastewater Treatment Plants do not properly fit the particularities of HSCW and provide little insight in the domain of HSCW. ...
English: The goal of this thesis is to study the viability of an IDSS based on a LFAC. At the end... more English: The goal of this thesis is to study the viability of an IDSS based on a LFAC. At the end, we intend to develop a complete IDSS application based on a LFAC which implements a real domain and which can be put to use in real scenarios. This thesis will focus on the problems combining two disciplines as di erent as IDSS and LFAC in order to make them work autonomously and transparently. We will work with the domain of a HSCW, a very common
Uploads
Papers by Dario Garcia-Gasulla