Corinne Amel Zayani

A user profile modelling using social annotations: a survey

As social networks are growing in terms of the number of users, resources and interactions; the user may be lost or unable to find useful information. Social elements could avoid this disorientation like the social annotations (tags) which become more and more popular and contribute to avoid the disorientation of the user. Representing a user based on these social annotations has showed their utility in reflecting an accurate user profile which could be used for a recommendation purpose. In this paper, we give a state of the art of characteristics of social user and techniques which model and update a tag-based profile. We show how to treat social annotations and the utility of modelling tag-based profiles for recommendation purposes.

Social collaborative service recommendation approach based on user’s trust and domain-specific expertise

A few years ago, the Internet of (Web) Service vision came to offer services to all aspects of life and business. The increasing number of Web services make service recommendation a directive research to help users discover services. Furthermore, the rapid development of social network has accelerated the development of social recommendation approach to avoid the data sparsity and cold-start problems that are not treated very well in the collaborative filtering approach. On the one hand, the pervasive use of the social media provides a big social information about the users (e.g., personnel data, social activities, relationships). Hence, the use of trust relation becomes a necessity to filter and select only the useful information. Several trust-aware service recommender systems have been proposed in literature but they do not consider the time in trust level detection among users. On the other hand, in the reality, the majority of users prefer the advice not only of their trusted friends but also their expertise in some domain-specific. In fact, the taking into account of user’ s expertise in recommendation step can resolve the user’ s disorientation problem. For these reasons, we present, in this paper, a Web service decentralized discovery approach which is based on two complementary mechanisms. The trust detection is the first mechanism to detect the social trust level among users. This level is defined in terms of the users’ interactions for a period of time and their interest similarity which are inferred from their social profiles. The service recommendation is the second mechanism which combines the social and collaborative approaches to recommend to the active user the appropriate services according to the expertise level of his most trustworthy friends. This level is extracted from the friends’ past invocation histories according to the domain-specific which is known in advance in the target user’s query. Performance evaluation shows that each proposed mechanism achieves good results. The proposed Level of social Trust (LoT) metric gives better precision more than 50% by comparing with the same metric without taking into account the time factor. The proposed service recommendation mechanism which based on the trust and the domain-specific expertise gives, firstly, a RMSE value lower than other trust-aware recommender systems like TidalTrust, MoleTrust and TrustWalker.Secondly, it provides a better response rate than the recommendation mechanism which based only on trust with a difference equal to 4%.

An Adaptive Method for User Profile Learning

The user profile is a key element in several systems which provide adapted result to the user. Thus, for a better quality of response and to satisfy the user, the profiles content must always be pertinent. So, the removal of irrelevant content is necessary. In this way, we propose in this paper a semi-supervised learning based method for automatically identifying irrelevant profile elements. The originality of this method is that it is based on a new co-training algorithm which is adapted to the content of any profile. For this, our method includes a preparation data step and a classification profile elements process. A comparative evaluation by the classical co-training algorithm shows that our method is better.

An Adaptation Approach: Query Enrichment by User Profile

In semi-structured information systems, generally, the adaptation of documents is essential to give the user the feeling that the query result is adapted to his preferences. The users needs can be defined in a user profile. But, in the literature, adaptation systems are designed for a particular domain and are oriented towards either navigation adaptation or content adaptation. Adaptation takes place after the users query has been evaluated. So, in this paper, we contribute to propose an adaptation algorithm which is domain independent and whose adaptation takes place before users query evaluation. This algorithm consists in enriching the user query on the basis of user profile in order to adapt the results to the user.

An Adaptive Navigation Method for Semi-structured Data

The navigation adaptation is the solution that supports the user during his interaction with the system. In the literature, several works that deal with the navigation adaptation are proposed. They guide the user from a document to another, provide the user with a set of links leading to the pertinent documents, or apply on simple links the suitable adaptive navigation technologies. In this paper, we contribute to propose a method that identifies the best navigation path between semi-structured result documents according to the user’s needs, history and device.

Pertinent User Profile based on Adaptive Semi-supervised Learning

Abstract Several systems such as adaptive systems, etc. provide responses to the user by taking into account, among other, his profile. After each user-system interaction, new information should be added to the user profile content. By the time and after several updating operations, the profile can become overloaded and the removal of irrelevant content is necessary. In this paper, we tackle the profile overloading problem. We propose a new method based on co-training algorithm for detecting and removing irrelevant elements. Our method is automatically adapted to the content of any profile and allows us to obtain the most generic classifier to each one. An experimental study by qualitative and comparative evaluations shows that the proposed method can detect and remove irrelevant profile content effectively.

A Case Study on the Influence of the User Profile Enrichment on Buzz Propagation in Social Media: Experiments on Delicious

The user is the main contributor for creating information in social media. In these media, users are influenced by the information shared through thenetwork. In a social context, there are so-called “buzz”, which is a technique to make noise around an event. This technique engenders that several users will be interested in this event at a time t. A buzz is then popular information in a specific time. A buzz may be a fact (true information) or a rumour (fake, false information). We are interested in studying buzz propagation through time in the social network Delicious. Also, we study the influence of enriched user profilesthat we proposed [2] to propagate the buzz in the same social network. In this paper, we state a case study on some information of the social network Delicious. This latter contains social annotations (tags) provided by users. These tags contribute to influence the other users to follow this information or to use it. This study relies onthree main axes: 1) we focus on tags considered as buzz and analyse their propagation through time 2) we consider a user profile as the set of tags provided by him. We will use the result of our previous work on dynamic user profile enrichment in order to analyse the influence of this enrichment in the buzz propagation. 3) we analyse each enriched user profile in order to show if the enrichment approach anticipate the buzz propagation. So, we can see the interest of filtering the information in order to avoid potential rumours and then, to propose relevant results to the user (e.g. avoid “bad” recommendation).

Trust Management in Social Internet of Things: A Survey

Social Internet of Things is a new paradigm where Internet of Things merges with Social Networks, allowing people and devices to interact, facilitating information sharing and enabling a variety of at-tractive applications. However, face to this new paradigm, users remain suspicious and careful. They fear disclosure of their data and violation of their privacy. Without trustworthy technologies to ensure user’s safe communications and trustworthy interactions, the SIoT will not reach enough popularity to be considered as a well-established technology. Accordingly, trust management becomes a major challenge to ensure reliable data analysis, qualified services and enhanced security. It helps people exceed their fears and promotes their acceptance and consumption on IoT services. However, current research still lacks a comprehensive study on trust management in SIoT. In this paper, we expose basic concepts, properties and models proposed for the trust management in SIOT environments. Furthermore, we discuss unsolved issues and future research trends.

A New Mashup Based Method for Event Detection from Social Media

Some events, such as terrorism attacks, earthquakes, and other events that represent tipping points, remain engraved in our memories. Today, through social media, researchers attempt to propose approaches for event detection. However, they are confronted to certain challenges owing to the noise of data propagated throughout social media. In this paper, a new mashup based method for event detection from social media is proposed using hadoop framework. The suggested approach aims at detecting real-world events by exploiting data collected from different social media sites. Indeed, the detected events are characterized by such descriptive dimensions as topic, time and location. Moreover, our approach assures a bilingual event detection. In fact, the proposed approach is able to detect events in English and French languages. In addition, our approach provides a mashup based multidimensional visualization by combining different multimedia components so as to add more details to the detected events. Furthermore, in order to overcome the problems occurring from the processing of big data, we integrated our approach into the hadoop distributed system.

A New Social Media Mashup Approach

Social media present a way to discover, report and share different types of events. For this reason, they are considered as a dynamic source of information which enables individuals to stay informed of all real-world events. This specificity encourages researchers to propose methods and approaches to detect events from social media. However many researchers are faced by divers challenges. This is due to the noise of data within the social media. In this paper, we propose a new approach for data retrieval and event detection, which differs from the previous ones by using the mashup concept. The proposed approach is able on the one hand to retrieve data from various social medias which have different structures. On the other hand, our approach aim to detect events with three main dimensions such as the topic, the time and the location.