Semantic User Interaction Profiles for Better People Recommendation

Stan, Johann; Do, Viet‐Hung; Maret, Pierre

doi:10.1109/asonam.2011.21

Cited by 17 publications

(7 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nonetheless, they does not scale well, because reasoning over logic fragments involved in the Semantic Web is, even for minimalist fragments like [29] in polynomial time, and at worst NEXPTIME [39]. Regarding the user recommendation process that follows user profile construction, [38] proposed the use of semantic technologies for better people recommendation in a system called Social Adviser. The authors introduced linked data (DBPedia, a semantized version of Wikipedia) in the content extraction process.…”

Section: Semantic-based Modelsmentioning

confidence: 99%

Real-time, scalable, content-based Twitter users recommendation

Subercaze

Gravier

Laforest

2016

WEB

View full text Add to dashboard Cite

Real-time recommendation of Twitter users based on the content of their profiles is a very challenging task. Traditional IR methods such as TF-IDF fail to handle efficiently large datasets. In this paper we present a scalable approach that allows real time recommendation of users based on their tweets. Our model builds a graph of terms, driven by the fact that users sharing similar interests will share similar terms. We show how this model can be encoded as a compact binary footprint, that allows very fast comparison and ranking, taking full advantage of modern CPU architectures. We validate our approach through an empirical evaluation against the Apache Lucene's implementation of TF-IDF. We show that our approach is in average two hundred times faster than standard optimized implementation of TF-IDF with a precision of 58 %.

show abstract

Section: Semantic-based Modelsmentioning

confidence: 99%

Real-time, scalable, content-based Twitter users recommendation

Subercaze

Gravier

Laforest

2016

WEB

View full text Add to dashboard Cite

show abstract

“…For instance, Zoltan and Johan [13] propose a system for the extraction of ontological topic concepts from tweets, topics are then weighted by their importance to the expert profile. Analysis of persistency of topics [14] as well as the awareness of profile dynamics [15,16] have been shown to improve user profiling, over static approaches. Twitter lists have been shown to be a rich source for user profiling [17].…”

Section: Expert Findingmentioning

confidence: 99%

“…Our approach favours the broader view of topics in accordance with the need for laterality and inclusion of people with borderline interest/expertise in a particular topic, who might bring innovative perspectives to the problem solving process -essential for our Open Innovation scenario. In cases where it is necessary to assure that only topics for which the user is really experienced about are represented, it is possible to use a more restrictive approach such as one of those proposed in [14], [15], and [16], mostly making use of the dynamics and persistence of topics in user's tweets.…”

Section: User Profiling On Twittermentioning

confidence: 99%

Finding Co-solvers on Twitter, with a Little Help from Linked Data

Stanković¹,

Rowe

Laublet³

2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In this paper we propose a method for suggesting potential collaborators for solving innovation challenges online, based on their competence, similarity of interests and social proximity with the user. We rely on Linked Data to derive a measure of semantic relatedness that we use to enrich both user profiles and innovation problems with additional relevant topics, thereby improving the performance of co-solver recommendation. We evaluate this approach against state of the art methods for query enrichment based on the distribution of topics in user profiles, and demonstrate its usefulness in recommending collaborators that are both complementary in competence and compatible with the user. Our experiments are grounded using data from the social networking service Twitter.com.

show abstract

“…Company emails are a rich source for learning more about each employee expertise and interests, but there may privacy and security concerns. Another, more acceptable source for such an RS may be represented by content employees share on internal or web-based social networks, such as Twitter [21] or Yammer. Such content is shorter and generally does not contain confidential information.…”

Section: Recommender Systems In the Enterprisementioning

confidence: 99%

Recommender Systems Using Social Network Analysis: Challenges and Future Trends

Stan

Muhlenbach

Largeron

2014

Encyclopedia of Social Network Analysis and Mining

Self Cite

View full text Add to dashboard Cite

Recommendation systems, Information filtering, Collaborative filtering, Content-based filtering GlossaryRecommender System (RS): Special type of information filtering system that provides a prediction that assists the user in evaluating items from a large collection that the user is likely to find interesting or useful.Status update (micropost): Short message, shared in an online social platform, expressing an activity, state of mind or opinion.Folksonomy: Whole set of tags that constitutes an unstructured collaborative knowledge classification scheme in a social tagging system. what online news to read [19], or, in the social network domain, which user to connect to or which users to consider as a trustful adviser. Overview Main Components of a Web 2.0 Social NetworkA social network can be defined as a set of entities interconnected and it is usually represented as a graph where the entities are described by nodes and their relationships by links. It should be noticed that this concept is not limited to the case of online social networks such as Facebook, LinkedIn, MySpace or Twitter, the main focus of our work.A common characteristic of these networks, and more specifically modern online social networks, is that they are composed of (i) users (with a user profile, activities and connections) and (ii) social objects representing the intermediations, e.g. topics of user interactions, shared videos, photos.The user profile generally includes static personal information, such as the name, email and address, as well as more dynamic information about the interests and information needs of the user. The role of the user profile is essential in online communities.Generally user profiles are different from one application to another, as users present themselves differently, based on the targeted population of the given application (which are sometimes very specific). Another dimension of users is represented by the activities they perform in the social platform. This includes content sharing, media uploading 3 and content description (such as photo tagging). Finally, the third dimension of users is represented by the social connections they establish with others in the network. Users in these online networks are generally connected to different communities, belonging to different social spheres (e.g., friends, family, coworkers).Another important user characteristic is related to trust. Indeed, the different applications on social content sites allow users to be closer to their communities and to be aware of peer activities and opinions. This brings new dimensions to trust and allows users to have higher confidence in the recommendations, suggestions and sentiment of friends.Shared social objects influence interactions between users. An object in this context has a concrete and perceptible, physical and/or numeric, manifestation. Some objects are the source of conversational interactions and keepers of collective attention.They constitute a conversation support. In our actual digital context objects are mainly multimed...

show abstract

Semantic User Interaction Profiles for Better People Recommendation

Cited by 17 publications

References 5 publications

Real-time, scalable, content-based Twitter users recommendation

Real-time, scalable, content-based Twitter users recommendation

Finding Co-solvers on Twitter, with a Little Help from Linked Data

Recommender Systems Using Social Network Analysis: Challenges and Future Trends

Contact Info

Product

Resources

About