Search

The aim of recommender systems is to provide users with items that could be of their interest. However one of the biggest drawbacks from recommender systems is the so called cold start problem, which occurs when new users or products are added to the system and therefore there is no previous information about them. There are many proposals in the literature that aim to deal with this issue. In some cases the user is required to provide some explicit information about them, which demands some effort on their part. Because of that and due to the great boom of social networks, we will focus on extracting implicit information from user’s social stream. In this paper we will present an approach on which social media data will be used to create a behavioural profile to classify the users and based on this classification will create predictions making use of machine learning techniques such as classification trees and random forests. Thus the user will not have to provide actively any kind of data explicitly but their social media source, alleviating in this way the cold start problem since the system would use this data in order to create user profiles, which will be the input for the engine of the recommender systems. We have carried out numerous experiments, as well as a comparison with some other state-of-the-art new user cold-start algorithms, obtaining very satisfactory results.

Recommender systems are tools that help users in the decision-making process of choosing items that may be relevant for them among a vast amount of other items. One of the main problems of recommender systems is the cold start problem, which occurs when either new items or new users are added to the system and, therefore, there is no previous information about them. This article presents a multi-source dataset optimized for the study and the alleviation of the cold start problem. This dataset contains info about the users, the items (movies), and ratings with some contextual information. The article also presents an example user behavior-driven algorithm using the introduced dataset for creating recommendations under the cold start situation. In order to create these recommendations, a mixed method using collaborative filtering and user-item classification has been proposed. The results show recommendations with high accuracy and prove the dataset to be a very good asset for future research in the field of recommender systems in general and with the cold start problem in particular., Spanish Government PID2019-103880RB-I00, Andalusian Agency project P20_00673

The rapid advances in Web technologies are promoting the development of new pedagogic models based on virtual teaching. In this framework, personalized services are necessary. Recommender systems can be used in an academic environment to assist users in their teaching-learning processes. In this paper, we present a trust based recommender system, adopting a fuzzy linguistic modeling, that provides personalized activities to students in order to reinforce their education, and applied it in the field of oral surgery and implantology. We don’t take into account users with similar ratings history but users in which each user can trust and we provide a method to aggregate the trust information. This system can be used in order to aid professors to provide students with a personalized monitoring of their studies with less effort. The results obtained in the experiments proved to be satisfactory., TIN2016-75850-R

A distributed database is a collection of data stored in different locations of a distributed system. The processing of queries in distributed databases is quite complex but of great importance for information management. Students who have to learn that process have serious difficulties for understanding them. On this work we present a web platform for helping the students learning the processing and optimization of queries in distributed databases. The novelty of this platform is that as far as we know, there is no similar graphical tool. It allows to visualize step by step the different phases of distributed query processing, showing how are they forming, making it easier for the students to understand these concepts. Moreover, having this web platform available, always and everywhere, indirectly have an impact on other competences like encouraging students’ autonomous work and self-learning, adapting the teaching to its one-time necessities and reinforcing the advantages to apply information techniques in the teaching field. The results of the developed tests to validate the platform's functionalities and student's satisfaction were very positive., This work has been developed thanks to the funding of the project PID46-201617 of the Universidad de Jaén.

Social Media (SM) has become the easiest, cheapest and fastest channel for companies to identify the events that affect their customers. The geo-location capabilities of the SM interactions enable Early Warning Systems to alert not only when the quality of service decays, but also where and how many customers are impacted. In this paper we present a system and a set of supporting metrics that exploit the geo-localized SM stream, quantify the perceived impact of events, incidents, etc. on a particular area over time. Industrial service providers can add this perceptional perspective to their standard monitoring tools to enable a prompt and appropriate reaction, the decision-making in marketing activities and to unveil customer acquisition opportunities applying the system to the competitors’ customers.

The job market is rapidly changing. Artificial Intelligence and automation technologies are reshaping the career market. Everyday, new jobs appear and new skills are added to the scope of existing job profiles. At the same time, some skills that once were assumed to be "must-haves" for particular jobs are no longer requested and some jobs are even becoming obsolete. The speed of changes as well as the increasing complexity of the job market introduce a key new challenge: there is no clear definition for a particular job in terms of skills and scope and consequently, people holding the same job title cannot be assumed to be actually doing the same thing. In addition, applicants find difficult to develop career paths, as the mapping of skills to particular jobs are fuzzier than ever before. In this article, we present a novel approach to homogenize the job definition, gathering first subject matter expertise using semantic expansion techniques on collaborative wikies, applying a word embeddings supported method to mine the skills from existing job posts and finally executing a semantic matching algorithm to converge to a consistent skills mapping. In order to show how our method performs, we apply it to one of the most popular, yet heterogeneous modern jobs, the data scientist and discuss the results obtained for the English speaking market., This paper has been developed with the FEDER financing under Project TIN2016-75850-R

One of the most difficult tasks for hedge funds investors is selecting a proper fund with just the right level level of risk. Often times, the issue is not only quantifying the hedge fund risk, but also the level the investors consider just right. To support this decision, we propose a novel recommender system, which is aware of the risks associated to different hedge funds, considering multiple factors, such as current yields, historic performance, diversification by industry, etc. Our system captures the preferences of the investors (e.g. industries, desired level of risk) applying fuzzy linguistic modeling and provides personalized recommendations for matching hedge funds. To demonstrate how our approach works, we have first profiled more than 4000 top hedge funds based on their composition and performance and second, created different simulated investment profiles and tested our recommendations with them., This paper has been developed with the FEDER financing under Project TIN2016-75850-R.

A method to quantify the impact of a topic over time in a location is proposed.We adopt the Recency, Frequency and Monetary model to ease the application.We introduce the concept of Social Media Engagement and Exposure to a topic.Our system harvests and processes geo-located Twitter to create the impact metrics.For the evaluation, we discussed the application to different real-world topics. Social Media can be used as a thermometer to measure how society perceives different news and topics. With the advent of mobile devices, users can interact with Social Media platforms anytime/anywhere, increasing the proportion of geo-located Social Media interactions and opening new doors to localized insights. This article suggests a new method built upon the industry standard Recency, Frequency and Monetary model to quantify the impact of a topic on a defined geographical location during a given period of time. We model each component with a set of metrics analyzing how users in the location actively engage with the topic and how they are exposed to the interactions in their Social Media network related to the topic. Our method implements a full fledged information extraction system consuming geo-localized Social Media interactions and generating on a regular basis the impact quantification metrics. To validate our approach, we analyze its performance in two real-world cases using geo-located tweets.

The enabling of geo-localization for Social Media content opens the door to a new set of applications based on the voice of the customer. For any company it is critical to understand both their own and their competitors’ strengths and weaknesses in all locations where they offer a service. With this motivation we created a Customers Acquisition and REtention system based on SOcial MEdia (CARESOME). Our system extracts and separates all social media interactions in a given location by market player and communication purpose and quantifies the impact of each single interaction over a given time period. To model the impact of the social media interactions, CARESOME relies on a set of metrics based on both intrinsic and extrinsic components—including Entity Engagement Index, Differential Perception Factor, Tie-Strength and Number of Exposed users—. In addition to the definition of our impact quantification metrics, we provide a thorough discussion about the design decisions taken to build our system. To illustrate the behavior of our system, we show-case a real world scenario from the airline industry based on two major airports in Great Britain.

Recommender systems (RSs) exploit past behaviors and user similarities to provide personalized recommendations. There are some precedents of usage in academic environments to assist users finding relevant information, based on assumptions about the characteristics of the items and users. Even if quality has already been taken into account as a property of items in previous works, it has never been given a key role in the re-ranking process for both items and users. In this paper, we present REFORE, a quality-based fuzzy linguistic REcommender system FOr REsearchers. We propose the use of some bibliometric measures as the way to quantify the quality of both items and users without the interaction of experts as well as the use of 2-tuple linguistic approach to describe the linguistic information. The system takes into account the measured quality as the main factor for the re-ranking of the top-N recommendations list in order to point out researchers to the latest and the best papers in their research fileds. To prove the accuracy improvement, we conduct a study involving different recommendation approaches, aiming at measuring their performance gain. The results obtained proved to be satisfactory for the researchers from different departments who took part on the tests.

Social Media is increasingly becoming an integral part of our lives and a place where an ever growing portion of our daily communication takes place. As we communicate, we reveal our emotions and this emotional chronicle is kept in our Social Media history. As the access to Internet became more pervasive, Social Media platforms could also store the location where the interactions took place, enabling the analysis of the emotions in these locations. Pursuing this idea, we suggest a method to create the emotional profile of a location based on the long-term emotional rating of the geo-localized SM interactions. In this paper we present our method based on a multivariate kernel density function of SM interactions on a Russell’s inspired circumplex plane, explain how we extract the emotions from Social Media Interactions relying on a modified version of extended Affective Norms for English Words and validate our approach with real-life locations.

The polarity detection problem typically relies on experimental dictionaries, where terms are assigned polarity scores lacking contextual information. As a matter of fact, the polarity is highly dependant on the domain or community it is analysed, so we can speak of a contextual bias. We propose a method supported by fuzzy linguistic modelling to quantify this contextual bias and to enable the bias-aware sentiment analysis. To show how our approach work, we measure the bias of common concepts in two different domains and discuss the results.

Every recommender system approach suffers the cold start problem to a greater or lesser extent. To soften this impact, the more common solution is to find the way of populating users profiles either using hybrid approach or finding external data sources. In this paper, we present a fuzzy linguistic approach that using bibliometrics aids to soft or remove the necessity of interaction of users providing them with personalized profiles built beforehand, thus reducing the cold start problem. To prove the effectiveness of the system, we conduct a test involving some researchers, aiming to build their profiles automatically. The results obtained proved to be satisfactory for the researchers.

The significant advances in information and communication technologies are changing the process of how information is accessed. The internet is a very important source of information and it influences the development of other media. Furthermore, the growth of digital content is a big problem for academic digital libraries, so that similar tools can be applied in this scope to provide users with access to the information. Given the importance of this, we have reviewed and analyzed several proposals that improve the processes of disseminating information in these university digital libraries and that promote access to information of interest. These proposals manage to adapt a user’s access to information according to his or her needs and preferences. As seen in the literature one of the techniques with the best results, is the application of recommender systems. These are tools whose objective is to evaluate and filter the vast amount of digital information that is accessible online in order to help users in their processes of accessing information. In particular, we are focused on the analysis of the fuzzy linguistic recommender systems (i.e., recommender systems that use fuzzy linguistic modeling tools to manage the user’s preferences and the uncertainty of the system in a qualitative way). Thus, in this work, we analyzed some proposals based on fuzzy linguistic recommender systems to help researchers, students, and teachers access resources of interest and thus, improve and complement the services provided by academic digital libraries., TIN2016-75850-R, TIN2013-40658-P

The latest development in cognitive technologies are helping us understand emotions and sentiments with unprecedented precision. Polarity detection is the key enabler to sentiment analysis and typically relies on experimental dictionaries, where terms are assigned polarity scores, yet lacking contextual information and based on human inputs and conventions. In this article, we present a novel approach to automatically extract a polarity dictionary from a particular domain, the stock market, without human intervention and addressing the scaling and thresholding problem. Our approach tracks the price changes of particular stocks over time, using it as a guiding polarity value. The magnitude of the price variation for a particular stock is then attributed to the financial news about this stock in corresponding period of time and that is what we use as our working corpus. On top of that, we derive the so-called binned corpus and apply the well-known TF–IDF information retrieval techniques to compute the TF–IDF value for each term. These values are then disseminated within the neighbourhood of each term based on the embeddings-enabled cosine distance. After introducing the problem and providing the background information, we thoroughly describe our method and all the components required to implement the system. Last but not least, we assign the terms to fuzzy linguistic labels and provide a volatility metric indicating how reliable our scores are depending on their distribution of occurrences in the corpus. To show how our approach works, we implement it for the Euro Stoxx 50 from January 2018 to March 2019 and discuss the results compared with typical approaches, pointing out potential improvements for further research work.

Low back pain affects a large proportion of the adult population at some point in their lives and has a major economic and social impact. To soften this impact, one possible solution is to make use of Information and Communication Technologies. Recommender systems, which exploit past behaviors and user similarities to predict possible user needs, have already been introduced in several health fields. In this paper, we present TPLUFIB-WEB, a fuzzy linguistic Web system that uses a recommender system to provide personalized exercises to patients with low back pain problems and to offer recommendations for their prevention. This system may be useful to reduce the economic impact of low back pain, help professionals to assist patients, and inform users on low back pain prevention measures. TPLUFIB-WEB satisfies the Web quality standards proposed by the Health On the Net Foundation (HON), Official College of Physicians of Barcelona, and Health Quality Agency of the Andalusian Regional Government, endorsing the health information provided and warranting the trust of users.

Recommender systems evaluate and filter the great amount of information available on the Web, so they could be used in an academic environment to help users in their searches of relevant information. In the literature, a lot of approaches for generating personalized recommendations of information items in such environment can be found. Usually, these approaches use user profiles and/or features of items to predict those relevant items, but they do not take into account the quality of these items. To overcome this problem, in this paper we propose a new recommender system based on quality. This system uses the quality of the items to estimate their relevance. The system measures the item quality and takes into account this measure like a new factor to be considered in the recommendation process. In such a way, we present a recommender system based on items' quality, to help users to access relevant research resources. This recommender systems is developed by using a fuzzy linguistic approach and it has been tested satisfactorily in a university digital library.

Recommender systems can be used in an academic environment to assist users in their decision making processes to find relevant information. In the literature we can find proposals based in user’ profile or in item’ profile, however they do not take into account the quality of items. In this work we propose the combination of item’ relevance for a user with its quality in order to generate more profitable and accurate recommendations. The system measures item quality and takes it into account as new factor in the recommendation process. We have developed the system adopting a fuzzy linguistic approach., Projects TIN2010-17876, TIC5299 y TIC-5991

Social Media is more and more positioning as the channel to implement early warning systems for prompt reaction to events that might affect a company’s core business. The enabling of geo-location for social media interactions unlocks new possibilities for decision making scenarios aiming at triage situation where the quality of service decays. In this work we provide a system and a set of soft metrics to quantify the impact created by the reaction of people directly affected by an incident in a particular area in order to facilitate the service providers’ appropriate reaction, the decision making in marketing activities and to unveil customer acquisition opportunities applying the system to the competitors’ customers., TIN2013-40658-P, TIN2016-75850-R

Recommender systems could be used to help users in their access processes to relevant information. Hybrid recommender systems represent a promising solution for multiple applications. In this paper we propose a hybrid fuzzy linguistic recommender system to help the Technology Transfer Office staff in the dissemination of research resources interesting for the users. The system recommends users both specialized and complementary research resources and additionally, it discovers potential collaboration possibilities in order to form multidisciplinary working groups. Thus, this system becomes an application that can be used to help the Technology Transfer Office staff to selectively disseminate the research knowledge and to increase its information discovering properties and personalization capacities in an academic environment.

We present a quality-based fuzzy linguistic recommender system for researchers. We propose the use of some bibliometrics measures as the way to quantify the quality of both items and users. The system takes into account the measured quality as the main factor for the re-ranking of the top-N recommendations list in order to point researchers to the latest and the best papers in their domain. To prove the accuracy improvement, we conduct a study involving dierent recommendation approaches. The results obtained proved to be satisfactory within the research departments who took part on the tests.

Information Technology and Quantitative Management (ITQM 2015), The seek of a personalized and quality education is the objective of Bologna process, but to carry out this task has a major economic impact. To soften this impact, one possible solution is to make use of recommender systems, which have already been introduced in several academic fields. In this paper, we present AyudasCBI, a novel fuzzy linguistic Web system that uses a recommender system to provide personalized activities to students to reinforce their individualized education. This system can be used in order to aid professors to provide students with a personalized monitoring of their studies with less effort. To prove the system, we conduct a study involving some students, aiming at measuring their performance. The results obtained proved to be satisfactory compared with the rest of the students who did not take part of the study., Projects TIC-5991 and TIN2013-40658-P

Information Technology and Quantitative Management (ITQM 2015), Recommender systems can be used to assist users in the process of accessing to relevant information. In the literature we can find sundry approaches for generating personalized recommendations and all of them make use of different users’ and/or items’ features. Building accurate profiles plays an essential role in this context, so that the system's success depend to a large extent on the ability of the learned profiles to represent the user's preferences. An ontology works very well to characterize the users profiles. In this paper we develop an ontology to characterize the trust between users using the fuzzy linguistic modelling, this way in the recommendation generation process we do not take into account users with similar ratings history but users in which each user can trust. We present our ontology and provide a method to aggregate the trust information captured in the trust-ontology and to update the user profiles based on the feedback., Projects UJA2013/08/41, TIN2013-40658-P, TIC5299, TIC-5991, TIN2012-36951 co-financed by FEDER and TIC6109

Low back pain affects a large proportion of the adult population at some point in their lives and has a major economic and social impact. To soften this impact, one possible solution is to make use of recommender systems, which have already been introduced in several health fields. In this paper, we present TPLUFIB-WEB, a novel fuzzy linguistic Web system that uses a recommender system to provide personalized exercises to patients with low back pain problems and to offer recommendations for their prevention. This system may be useful to reduce the economic impact of low back pain, help professionals to assist patients, and inform users on low back pain prevention measures. A strong part of TPLUFIB-WEB is that it satisfies the Web quality standards proposed by the Health On the Net Foundation (HON), Official College of Physicians of Barcelona, and Health Quality Agency of the Andalusian Regional Government, endorsing the health information provided and warranting the trust of users.

Recommender systems evaluate and filter the great amount of information available on the Web, so they could be used to help users in their access processes to relevant information. In the literature we can find a lot of approaches for generating personalized recommendations. Hybrid recommender systems combine in different ways several approaches, so these recommendation strategies represent a promising solution for multiple applications. In this paper we propose a hybrid fuzzy linguistic recommender system to help the Technology Transfer Office staff in the dissemination of research resources interesting for the users. The system recommends users both specialized and complementary research resources and additionally, it discovers potential collaboration possibilities in order to form multidisciplinary working groups.

In the recent times the amount of information coming overwhelms us, and because of it we have serious problems to access to relevant information, that is, we suffer information overload problems. Recommender systems have been applied successfully to avoid the information overload in different scopes, but the number of electronic resources daily generated keeps growing and the problem still remain. Therefore, we find a persistent problem of information overload. In this paper we propose an improved recommender system to avoid the persistent information overload found in a University Digital Library. The idea is to include a memory to remember selected resources but not recommended to the user, and in such a way, the system could incorporate them in future recommendations to complete the set of filtered resources, for example, if there are a few resources to be recommended or if the user wishes output obtained by combination of resources selected in different recommendation rounds.