Abstract. In this paper we investigate two fundamental issues related to the production of cohere... more Abstract. In this paper we investigate two fundamental issues related to the production of coherent discourse by intelligent agents: a cohe-sion property and a fluency property. The cohesion aspects of discourse production relate to the use of pronominal anaphora: whether and in what conditions intelligent agents could acquire pronouns as means to express recently mentioned entities? We show that the acquisition of pro-nouns in the vocabulary of an agent is conditioned by the existence of a memory channel recording the object previously in focus. The approach follows an evolutionary paradigm of language acquisition. Experiments show that pronouns spontaneously appear in the vocabulary of a com-munity of 10 agents dialogging on a static scene and that, generally, the use of pronouns enhance the communication success. The processing load experiments address the fluency of discourse, measured in terms of Cen-tering transitions. Contrary to previous findings, this side of discourse cohe...
2020 22nd International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)
Crack identification is a common problem that requires human involvement and manual identificatio... more Crack identification is a common problem that requires human involvement and manual identification. Our work is focused on detecting street surface cracks using Computer Vision algorithms. The problem at hand has been divided in three steps: (i) the first step transforms a given 3D video in 2D individual frames, (ii) the second step processes these frames in order to identify the relevant part of the street using Support Vector Machine and Vanishing Point Detection and (iii) in the third step the detection itself has been implemented using three methods: Convolutional Neural Network, U-Net and a Local Binary Pattern. In this paper we present our methods, experiments and results for detecting cracks on surfaces like streets and sidewalks.
This article presents the results of an implementation of a new platform based on swarm communica... more This article presents the results of an implementation of a new platform based on swarm communication and executable choreographies. In our research of executable choreographies, we have come up with a more general model to implement smart contracts and a generic architecture of systems using hierarchical blockchain architecture. The novel concepts of secret smart contract and near-chain are introduced. The near-chain approach presents a new method to extend the hierarchical blockchain architecture and to improve performance, security and privacy characteristics of general blockchain-based systems. As such, we are subsequently defining and explaining why any extension of blockchain architectures should revolve around three essential dimensions: trustlessness, nonrepudiation and tamper resistance. The hierarchical blockchain approach provides a novel perspective, as well as establishing off-chain storages (near-chains) as special types of hierarchical blockchains stored in a distribu...
Augmented reality is used today in many different domains with important benefits for the end use... more Augmented reality is used today in many different domains with important benefits for the end users. They are helped to understand the concepts better or to do simple things guided by augmented information presented like texts, images, audios or videos. In this paper we present an augmented reality application created to help the user identify fast the existing ingredients in a fridge and then choose a recipe which can be prepared with these ingredients. The application also provides the necessary steps to prepare this recipe and suggestive images which can help the user during the preparation process. Author
In this paper, an application built by us with aim to provide users the possibility to explore to... more In this paper, an application built by us with aim to provide users the possibility to explore topics from Twitter and find out people’s opinion even if it is a positive one or a negative one is presented. The result of searched topic was divided into groups, based on the named entity from tweets. Using algorithms that calculate distance between two strings, similar tweets was removed. More than that, users have the opportunity to see a sentiment analysis of tweets. Author
103 Event detection in Tweets Andrei-Bogdan Baran “Alexandru Ioan Cuza” University, Faculty of Co... more 103 Event detection in Tweets Andrei-Bogdan Baran “Alexandru Ioan Cuza” University, Faculty of Computer Science General Berthelot, No. 16 andrei.baran@info.uaic.ro Adrian Iftene “Alexandru Ioan Cuza” University, Faculty of Computer Science General Berthelot, No. 16 adiftene@info.uaic.ro ABSTRACT Twitter is among the fastest-growing online social networking services, with more than 140 million users producing over 400 million tweets per day. It enables users to post status updates (tweets) about a huge variety of topics to a network of followers using various communication services such as cell phones, e-mails, Web interfaces, or other third-party applications. Monitoring and analyzing this rich and continuous usergenerated content can lead to obtaining valuable information about local and global news and events, because virtually, any person witnessing or involved in any event is nowadays able to disseminate realtime information, which can reach the other side of the world as the ev...
The current paper exposes the development of a game application that combines augmented reality, ... more The current paper exposes the development of a game application that combines augmented reality, modern robotics, mobile features and multiplayer networking in order to create an immersive collaboration system, where people can interconnect to have fun and complete entertaining tasks together. SMAUG is a game based on Sphero 2.0 robotic ball, where multiple players share the same driving control of the robot. Depending on the game session and quest, a player can influence a certain moving direction or the robot's speed. Furthermore, the playground is augmented on the driving surface, so each player is looking to the game scene through his/her mobile’s camera. The main objective here is that players need to collaborate in order to accomplish various tasks: collecting game artifacts, drive on augmented tracks, break down walls/gates, escape mazes and so on. In order to accomplish this objective the game is demanding collaboration: the game artifacts are distributed to different pl...
In the last years, multimedia content has grown increasingly over the Internet, especially in soc... more In the last years, multimedia content has grown increasingly over the Internet, especially in social networks (like Facebook or Flickr), where users often post images using their mobile devices. In these networks, the content is later used in search operations, when some users want to find something using a specific query. Nowadays, searching into these networks is primarily made using the title, description and the keywords associated to resources added by users that have posted the content. In this paper we address the problem of query ambiguity. The usage of semantic resources, like ConceptNet and DBpedia ontologies, has been proven to retrieve a better set of results.
Social networks such as Twitter or Facebook grew rapidly in popularity, and users use them to sha... more Social networks such as Twitter or Facebook grew rapidly in popularity, and users use them to share opinions about topics of interest, to be part of the community or to post messages that are available everywhere. This paper presents a system created in order to process streamed data taken from Twitter and classify it into positive, negative or neutral. The results of these processing’s can be visualized in a suggestive manner on Google Maps, users can select the language of the tweets, can group tweets that present the same news and can even display a dynamic evolution of the news in terms of its appearance. With all this amount of information it is very opportune to do some data analysis to detect different types of events (and their locations) that happen worldwide, especially at the time when this data represents information related to refugee crisis or signals terrorist attacks.
In Romania, stroke is the second leading cause of death and disability (Neagu, 2019). The speed o... more In Romania, stroke is the second leading cause of death and disability (Neagu, 2019). The speed of reaction in the case of a stroke can make the difference between life and death, between a healthy patient and a patient who remains disabled for life. In this context, we wanted to create a mobile application in Romanian that will help users to do a F.A.S.T. test very quickly. The test identifies possible problems at the face, arms and speech levels. This article shows how we made this application using a client-server architecture and what users think about it. Author
This paper addresses the increasingly encountered challenge of ontology alignment. Starting with ... more This paper addresses the increasingly encountered challenge of ontology alignment. Starting with basic similarity measures such as the syntactic similarity, represented by the Levenshtein or Jaro Distance, semantic similarities, which make use of WordNet and taxonomy similarities, our new system uses a genetic algorithm specially designed for the task of optimizing the aggregation of these measures. Assessment done by us in the last part of the paper demonstrates the usefulness of the genetic algorithm, which manage a consistent improvement of classical alignment methods.
Health care industry is one of the most interesting domains where we can bring innovative and sma... more Health care industry is one of the most interesting domains where we can bring innovative and smart technologies that can have a big positive impact on our health. These days people are starting to be more and more concerned about their health and they are interested in having small wearable devices that allow them to be able to get real time data through different sensors regarding the environment or different activities they are involved in. In this paper we will show how new disruptive technologies can be used to monitored the level of the ultraviolet radiation and then how the user can be notified about that. Author
In the last years, the volume of data (text, image or multi- media) has grown exponentially. In t... more In the last years, the volume of data (text, image or multi- media) has grown exponentially. In this context, the information retrieval task has become more and more difficult. This paper aims to present an application that performs image retrieval based on Flickr network image collection. After the user fills in the query, the system is responsible with (1) query processing, (2) image retrieval and (3) results displaying. Before retrieving the results, the system clusters the images based on several cri- terions and shows to the user one representative image for every cluster it creates. In this way, the results displayed to the user are diverse and he still has the possibility to view all the images from a cluster, when clicking on the representative image. The system achieves diversity in two ways: (i) at query processing, when the user adds from WordNet the lexical family corresponding to query keywords, and (ii) when the results are displayed and the clusters are created.
Every device that is released nowadays has the capability to display videos. From small devices l... more Every device that is released nowadays has the capability to display videos. From small devices like smart phones to bigger ones likes desktops or smart TVs, movies are available at any time. Therefore, getting important, fast and reliable data about movies at any time and everywhere is very important when it comes to money. Computer users like to get live experiences based on their interests, searches or their personal profile and their needs include movies. When people gather to someone’s place or a family unites on holidays, finding a good movie to watch for that moment becomes hard. There are different tastes in movies, persons that like drama more than science-fiction or some that will enjoy movies based on the director of that movie or important actors. Using a good search engine that might understand users desires based on recent searches or interests might come in handy. Movies Classification application runs on Windows operating systems and tries to understand user perspect...
Commonsense Validation and Explanation has been a difficult task for machines since the dawn of c... more Commonsense Validation and Explanation has been a difficult task for machines since the dawn of computing. Although very trivial to humans it poses a high complexity for machines due to the necessity of inference over a pre-existing knowledge base. In order to try and solve this problem the SemEval 2020 Task 4 - ”Commonsense Validation and Explanation (ComVE)” aims to evaluate systems capable of multiple stages of ComVE. The challenge includes 3 tasks (A, B and C), each with it’s own requirements. Our team participated only in task A which required selecting the statement that made the least sense. We choose to use a bidirectional transformer in order to solve the challenge, this paper presents the details of our method, runs and result.
This paper presents main activities of UAIC (“Alexandru Ioan Cuza” University) team from the MUCK... more This paper presents main activities of UAIC (“Alexandru Ioan Cuza” University) team from the MUCKE project. MUCKE addressed the stream of multimedia social data with new and reliable knowledge extraction models designed for multilingual and multimodal data shared on social networks. Credibility models for multimedia streams are a novel topic, and constituted the main scientific contribution of the project. UAIC group was involved in the main tasks of the project: building the data collection, text processing, diversification in image retrieval and data credibility. Author
In this paper, it is presented a system built with the aim to predict plant species given their l... more In this paper, it is presented a system built with the aim to predict plant species given their location in the context of the GeoLifeCLEF 2019 challenge. There are developed several prediction models on the basis of a representation in the environmental space (a feature vector composed of climatic variables and other variables such as soil type, land cover, distance to water) and using the spatial occurrences of species (latitude and longitude): Random Forest, KNearest Neighbors, Artificial Neural Networks and XGBoost.
One of the aims of this paper is to present a new programming paradigm based on the new paradigms... more One of the aims of this paper is to present a new programming paradigm based on the new paradigms intensively used in IT industry. Implementation of these techniques can improve the quality of code through modularization, not only in terms of entities used by a program, but also in terms of states in which they pass. Another aspect followed in this paper takes into account that in the development of software applications, the transition from the design to the source code is a very expensive step in terms of effort and time spent. Diagrams can hide very important details for simplicity of understanding, which can lead to incorrect or incomplete implementations. To improve this process communicative automaton based programming comes with an intermediate step. We will see how it goes after creating modeling diagrams to communicative automata and then to writing code for each of them. We show how the transition from one step to another is much easier and intuitive.
Abstract. In this paper we investigate two fundamental issues related to the production of cohere... more Abstract. In this paper we investigate two fundamental issues related to the production of coherent discourse by intelligent agents: a cohe-sion property and a fluency property. The cohesion aspects of discourse production relate to the use of pronominal anaphora: whether and in what conditions intelligent agents could acquire pronouns as means to express recently mentioned entities? We show that the acquisition of pro-nouns in the vocabulary of an agent is conditioned by the existence of a memory channel recording the object previously in focus. The approach follows an evolutionary paradigm of language acquisition. Experiments show that pronouns spontaneously appear in the vocabulary of a com-munity of 10 agents dialogging on a static scene and that, generally, the use of pronouns enhance the communication success. The processing load experiments address the fluency of discourse, measured in terms of Cen-tering transitions. Contrary to previous findings, this side of discourse cohe...
2020 22nd International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)
Crack identification is a common problem that requires human involvement and manual identificatio... more Crack identification is a common problem that requires human involvement and manual identification. Our work is focused on detecting street surface cracks using Computer Vision algorithms. The problem at hand has been divided in three steps: (i) the first step transforms a given 3D video in 2D individual frames, (ii) the second step processes these frames in order to identify the relevant part of the street using Support Vector Machine and Vanishing Point Detection and (iii) in the third step the detection itself has been implemented using three methods: Convolutional Neural Network, U-Net and a Local Binary Pattern. In this paper we present our methods, experiments and results for detecting cracks on surfaces like streets and sidewalks.
This article presents the results of an implementation of a new platform based on swarm communica... more This article presents the results of an implementation of a new platform based on swarm communication and executable choreographies. In our research of executable choreographies, we have come up with a more general model to implement smart contracts and a generic architecture of systems using hierarchical blockchain architecture. The novel concepts of secret smart contract and near-chain are introduced. The near-chain approach presents a new method to extend the hierarchical blockchain architecture and to improve performance, security and privacy characteristics of general blockchain-based systems. As such, we are subsequently defining and explaining why any extension of blockchain architectures should revolve around three essential dimensions: trustlessness, nonrepudiation and tamper resistance. The hierarchical blockchain approach provides a novel perspective, as well as establishing off-chain storages (near-chains) as special types of hierarchical blockchains stored in a distribu...
Augmented reality is used today in many different domains with important benefits for the end use... more Augmented reality is used today in many different domains with important benefits for the end users. They are helped to understand the concepts better or to do simple things guided by augmented information presented like texts, images, audios or videos. In this paper we present an augmented reality application created to help the user identify fast the existing ingredients in a fridge and then choose a recipe which can be prepared with these ingredients. The application also provides the necessary steps to prepare this recipe and suggestive images which can help the user during the preparation process. Author
In this paper, an application built by us with aim to provide users the possibility to explore to... more In this paper, an application built by us with aim to provide users the possibility to explore topics from Twitter and find out people’s opinion even if it is a positive one or a negative one is presented. The result of searched topic was divided into groups, based on the named entity from tweets. Using algorithms that calculate distance between two strings, similar tweets was removed. More than that, users have the opportunity to see a sentiment analysis of tweets. Author
103 Event detection in Tweets Andrei-Bogdan Baran “Alexandru Ioan Cuza” University, Faculty of Co... more 103 Event detection in Tweets Andrei-Bogdan Baran “Alexandru Ioan Cuza” University, Faculty of Computer Science General Berthelot, No. 16 andrei.baran@info.uaic.ro Adrian Iftene “Alexandru Ioan Cuza” University, Faculty of Computer Science General Berthelot, No. 16 adiftene@info.uaic.ro ABSTRACT Twitter is among the fastest-growing online social networking services, with more than 140 million users producing over 400 million tweets per day. It enables users to post status updates (tweets) about a huge variety of topics to a network of followers using various communication services such as cell phones, e-mails, Web interfaces, or other third-party applications. Monitoring and analyzing this rich and continuous usergenerated content can lead to obtaining valuable information about local and global news and events, because virtually, any person witnessing or involved in any event is nowadays able to disseminate realtime information, which can reach the other side of the world as the ev...
The current paper exposes the development of a game application that combines augmented reality, ... more The current paper exposes the development of a game application that combines augmented reality, modern robotics, mobile features and multiplayer networking in order to create an immersive collaboration system, where people can interconnect to have fun and complete entertaining tasks together. SMAUG is a game based on Sphero 2.0 robotic ball, where multiple players share the same driving control of the robot. Depending on the game session and quest, a player can influence a certain moving direction or the robot's speed. Furthermore, the playground is augmented on the driving surface, so each player is looking to the game scene through his/her mobile’s camera. The main objective here is that players need to collaborate in order to accomplish various tasks: collecting game artifacts, drive on augmented tracks, break down walls/gates, escape mazes and so on. In order to accomplish this objective the game is demanding collaboration: the game artifacts are distributed to different pl...
In the last years, multimedia content has grown increasingly over the Internet, especially in soc... more In the last years, multimedia content has grown increasingly over the Internet, especially in social networks (like Facebook or Flickr), where users often post images using their mobile devices. In these networks, the content is later used in search operations, when some users want to find something using a specific query. Nowadays, searching into these networks is primarily made using the title, description and the keywords associated to resources added by users that have posted the content. In this paper we address the problem of query ambiguity. The usage of semantic resources, like ConceptNet and DBpedia ontologies, has been proven to retrieve a better set of results.
Social networks such as Twitter or Facebook grew rapidly in popularity, and users use them to sha... more Social networks such as Twitter or Facebook grew rapidly in popularity, and users use them to share opinions about topics of interest, to be part of the community or to post messages that are available everywhere. This paper presents a system created in order to process streamed data taken from Twitter and classify it into positive, negative or neutral. The results of these processing’s can be visualized in a suggestive manner on Google Maps, users can select the language of the tweets, can group tweets that present the same news and can even display a dynamic evolution of the news in terms of its appearance. With all this amount of information it is very opportune to do some data analysis to detect different types of events (and their locations) that happen worldwide, especially at the time when this data represents information related to refugee crisis or signals terrorist attacks.
In Romania, stroke is the second leading cause of death and disability (Neagu, 2019). The speed o... more In Romania, stroke is the second leading cause of death and disability (Neagu, 2019). The speed of reaction in the case of a stroke can make the difference between life and death, between a healthy patient and a patient who remains disabled for life. In this context, we wanted to create a mobile application in Romanian that will help users to do a F.A.S.T. test very quickly. The test identifies possible problems at the face, arms and speech levels. This article shows how we made this application using a client-server architecture and what users think about it. Author
This paper addresses the increasingly encountered challenge of ontology alignment. Starting with ... more This paper addresses the increasingly encountered challenge of ontology alignment. Starting with basic similarity measures such as the syntactic similarity, represented by the Levenshtein or Jaro Distance, semantic similarities, which make use of WordNet and taxonomy similarities, our new system uses a genetic algorithm specially designed for the task of optimizing the aggregation of these measures. Assessment done by us in the last part of the paper demonstrates the usefulness of the genetic algorithm, which manage a consistent improvement of classical alignment methods.
Health care industry is one of the most interesting domains where we can bring innovative and sma... more Health care industry is one of the most interesting domains where we can bring innovative and smart technologies that can have a big positive impact on our health. These days people are starting to be more and more concerned about their health and they are interested in having small wearable devices that allow them to be able to get real time data through different sensors regarding the environment or different activities they are involved in. In this paper we will show how new disruptive technologies can be used to monitored the level of the ultraviolet radiation and then how the user can be notified about that. Author
In the last years, the volume of data (text, image or multi- media) has grown exponentially. In t... more In the last years, the volume of data (text, image or multi- media) has grown exponentially. In this context, the information retrieval task has become more and more difficult. This paper aims to present an application that performs image retrieval based on Flickr network image collection. After the user fills in the query, the system is responsible with (1) query processing, (2) image retrieval and (3) results displaying. Before retrieving the results, the system clusters the images based on several cri- terions and shows to the user one representative image for every cluster it creates. In this way, the results displayed to the user are diverse and he still has the possibility to view all the images from a cluster, when clicking on the representative image. The system achieves diversity in two ways: (i) at query processing, when the user adds from WordNet the lexical family corresponding to query keywords, and (ii) when the results are displayed and the clusters are created.
Every device that is released nowadays has the capability to display videos. From small devices l... more Every device that is released nowadays has the capability to display videos. From small devices like smart phones to bigger ones likes desktops or smart TVs, movies are available at any time. Therefore, getting important, fast and reliable data about movies at any time and everywhere is very important when it comes to money. Computer users like to get live experiences based on their interests, searches or their personal profile and their needs include movies. When people gather to someone’s place or a family unites on holidays, finding a good movie to watch for that moment becomes hard. There are different tastes in movies, persons that like drama more than science-fiction or some that will enjoy movies based on the director of that movie or important actors. Using a good search engine that might understand users desires based on recent searches or interests might come in handy. Movies Classification application runs on Windows operating systems and tries to understand user perspect...
Commonsense Validation and Explanation has been a difficult task for machines since the dawn of c... more Commonsense Validation and Explanation has been a difficult task for machines since the dawn of computing. Although very trivial to humans it poses a high complexity for machines due to the necessity of inference over a pre-existing knowledge base. In order to try and solve this problem the SemEval 2020 Task 4 - ”Commonsense Validation and Explanation (ComVE)” aims to evaluate systems capable of multiple stages of ComVE. The challenge includes 3 tasks (A, B and C), each with it’s own requirements. Our team participated only in task A which required selecting the statement that made the least sense. We choose to use a bidirectional transformer in order to solve the challenge, this paper presents the details of our method, runs and result.
This paper presents main activities of UAIC (“Alexandru Ioan Cuza” University) team from the MUCK... more This paper presents main activities of UAIC (“Alexandru Ioan Cuza” University) team from the MUCKE project. MUCKE addressed the stream of multimedia social data with new and reliable knowledge extraction models designed for multilingual and multimodal data shared on social networks. Credibility models for multimedia streams are a novel topic, and constituted the main scientific contribution of the project. UAIC group was involved in the main tasks of the project: building the data collection, text processing, diversification in image retrieval and data credibility. Author
In this paper, it is presented a system built with the aim to predict plant species given their l... more In this paper, it is presented a system built with the aim to predict plant species given their location in the context of the GeoLifeCLEF 2019 challenge. There are developed several prediction models on the basis of a representation in the environmental space (a feature vector composed of climatic variables and other variables such as soil type, land cover, distance to water) and using the spatial occurrences of species (latitude and longitude): Random Forest, KNearest Neighbors, Artificial Neural Networks and XGBoost.
One of the aims of this paper is to present a new programming paradigm based on the new paradigms... more One of the aims of this paper is to present a new programming paradigm based on the new paradigms intensively used in IT industry. Implementation of these techniques can improve the quality of code through modularization, not only in terms of entities used by a program, but also in terms of states in which they pass. Another aspect followed in this paper takes into account that in the development of software applications, the transition from the design to the source code is a very expensive step in terms of effort and time spent. Diagrams can hide very important details for simplicity of understanding, which can lead to incorrect or incomplete implementations. To improve this process communicative automaton based programming comes with an intermediate step. We will see how it goes after creating modeling diagrams to communicative automata and then to writing code for each of them. We show how the transition from one step to another is much easier and intuitive.
Uploads
Papers