scispace - formally typeset
Search or ask a question
Book ChapterDOI

Personalizing and improving tag-based search in folksonomies

TL;DR: The aim in this contribution is to propose new technique for the Social Semantic Web technologies in order to see how to overcome the problem of tags' ambiguity automatically in folksonomies even when the authors choose representing these latter with ontologies.
Abstract: Recently, the approaches that combine semantic web ontologies and web 2.0 technologies have constituted a significant research field. We present in this paper an original approach concerning a technology that has recognized a great popularity in these recent years, we talk about folksonomies. Our aim in this contribution is propose new technique for the Social Semantic Web technologies in order to see how we can overcome the problem of tags' ambiguity automatically in folksonomies even when we choose representing these latter with ontologies. We'll also illustrate how we can enrich any folksonomy by a set of pertinent data to improve and facilitate the resources' retrieval in these systems; all this with tackling another problem, we speak about spelling variations.

Summary (3 min read)

1 Introduction

  • Among the powerful technologies of Web 2.0, the authors find folksonomies, this term has recently appeared on the net to describe a system of classification derived from the practice and method of collaboratively creating and managing tags to annotate and categorize content.
  • Ontologies which constitute the backbone of semantic web contribute significantly in solving the problems of semantics during the definition and the search of information.
  • As examples the authors can cite the problem of tags’ ambiguity and spelling variations (or Synonymy) in folksonomies.
  • After in Section 4 the authors move to the experimental phase in order to measure the performance of their approach and discuss the obtained results.

3 Semantic Social Folksonomy with Ontology (SSFO)

  • The aim in this contribution is to introduce both the semantics and the social aspects in folksonomies in order to let any user in the system retrieving relevant web resources close to his preferences.
  • The authors aim to show how one can produce a technique for helping any ontology already used for representing a folksonomy to overcome the problem of tags’ ambiguity automatically without the need of an expert who must control and organize links between terms.
  • In addition the authors want show how they can enrich their folksonomy (without human intervention) with relevant data in order to help optimizing the time of search and enormously reduce the problems of spelling variations and the lack of semantics within folksonomies focusing on the rules-based systems.

3.2 Resolving Tags’ Ambiguity in Folksonomies

  • The authors technique to overcome the problem of tags’ ambiguity is not based on ontologies.
  • The idea is to study the profile of each member in the system and then compare the preferences of this one with other users in order to extract those who are similar to him.
  • To make the system flexible, the authors propose to make it interact with the user to accept or reject the retrieved resources.
  • And to avoid the "cold start" problem which is generally occur from a lack of the required data by the system in order to make an excellent recommendation; it’s proposed to measure the similarity between resources when the users are not similar.
  • To achieve this classification, the authors propose to calculate the ratio between the number of resources used by the user himself (i.e. the one who does the search) and the number of the resources shared between him and the other users.

3.3 Rules-Based Systems in Folksonomies

  • The purpose of using rules-based systems can be summarized as follow: 1) Avoid the existence of an expert who must control and organize links between terms.
  • But it will optimize this time and also the memory space that can be lost in each calculation because with this process; before their system begin the calculation of similarity between users or between resources it will firstly see in the fact base if there are resources similar to those already proposed to this user.
  • Furthermore the strength of this language is appear from the fact that it can support many kinds of dialects; among them the authors find the RIF-PRD (the Production Rule Dialect of the W3C Rule Interchange Format) [4] which allows adding, deleting and modifying facts in the fact base.

4.1 Dataset and Data Treatment

  • In order to validate their approach, the authors have conducted an experiment with del.icio.us database.
  • The authors test base comprises 1605 tag assignments involving 55 users, 526 tags some of which are ambiguous or have many spelling variations, 950 resources each having possibly several tags and several users.
  • This ambiguity of tags has been subjectively decided: for instance apple is ambiguous and software is not.
  • It should be noted that the authors have used a simple properties for describing this ontology in order to avoid losing the meaning and the objective of their approach, where they have suggested representing their folksonomy by a simple ontology defined by primitives relations such as "tagged by" and "used by"…etc.
  • After that the authors have used a tool for social network analysis called "Pajek1", in order to extract the three networks 'Users-Tags', 'Users-Resources' and 'TagsResources'.

4.2 Results

  • Three metrics are used for evaluating their approach: Precision:.
  • It measures the system's ability to reject all not relevant resources to a query.
  • It is given by the ratio of all relevant selected resources and the set of all selected resources.
  • It is given by the ratio of relevant retrieved resources and all relevant resources in the database.
  • The three metrics listed above are calculated for each user, and then the average of each metric in the system is calculated.

4.3 Discussion

  • The approach presented in this work has tried to extract the semantics in folksonomies in order to allow users capturing the social dimension of their tagging activity.
  • Indeed the obtained results show that the technique SSFO succeeded in distinguishing between ambiguous tags and also them which have many spelling variations.
  • In addition the work presented in [8] doesn’t tackle the problem of spelling variations.
  • In comparison with [5], the authors find that their approach doesn’t need an expert who must control and organize links between terms.
  • Also the expertise of users which was introduced in [6] is characterized by the complexity of its exploitation when the authors try as much as possible to avoid a cognitive overload, to limit the necessary effort for the formalization of this expertise which is achieved by their approach.

Did you find this useful? Give us your feedback

Content maybe subject to copyright    Report

HAL Id: hal-01201745
https://hal.inria.fr/hal-01201745
Submitted on 30 Apr 2016
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-
entic research documents, whether they are pub-
lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diusion de documents
scientiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Personalizing and Improving Tag-Based Search in
Folksonomies
Samia Beldjoudi, Hassina Seridi-Bouchelaghem, Catherine Faron Zucker
To cite this version:
Samia Beldjoudi, Hassina Seridi-Bouchelaghem, Catherine Faron Zucker. Personalizing and Improving
Tag-Based Search in Folksonomies. 15th International Conference on Articial Intelligence: Method-
ology, Systems, and Applications, AIMSA 2012, 2012, Varna, Bulgaria. �10.1007/978-3-642-33185-
5_12�. �hal-01201745�

Personalizing and Improving Tag-Based Search
in Folksonomies
Samia Beldjoudi
1
, Hassina Seridi-Bouchelaghem
1
, and Catherine Faron-Zucker
2
1
Laboratory of Electronic Document Management LabGED,
Badji Mokhtar University Annaba, Algeria
{beldjoudi,seridi}@labged.net
2
I3S, Université Nice - Sophia Antipolis, CNRS 930 route des Colles, BP 145,
06930 Sophia Antipolis Cedex, France
catherine.faron-zucker@unice.fr
Abstract. Recently, the approaches that combine semantic web ontologies and
web 2.0 technologies have constituted a significant research field. We present in
this paper an original approach concerning a technology that has recognized a
great popularity in these recent years, we talk about folksonomies. Our aim in
this contribution is propose new technique for the Social Semantic Web
technologies in order to see how we can overcome the problem of tags’
ambiguity automatically in folksonomies even when we choose representing
these latter with ontologies. We’ll also illustrate how we can enrich any
folksonomy by a set of pertinent data to improve and facilitate the resources’
retrieval in these systems; all this with tackling another problem, we speak
about spelling variations.
Keywords: Folksonomies, Web 2.0, Semantic Web, Tags Ambiguity, Spelling
Variations.
1 Introduction
Among the powerful technologies of Web 2.0, we find folksonomies, this term has
recently appeared on the net to describe a system of classification derived from the
practice and method of collaboratively creating and managing tags to annotate and
categorize content. Ontologies which constitute the backbone of semantic web
contribute significantly in solving the problems of semantics during the definition and
the search of information. However even with the strong points of folksonomies and
ontologies; their combination together still suffers from some problems. As examples
we can cite the problem of tags’ ambiguity and spelling variations (or Synonymy) in
folksonomies. Our goal in this contribution is to show how we can exploit the power
of social interactions between the folksonomy’s members in order to extract the
meaning of terms and overcome the problems of tags’ ambiguity and spelling
variations. Also we will try to show how we can use the principle of rules-based
systems with ontologies for helping our system to enhance automatically the
folksonomy by relevant facts can increase the data available within our system with

Personalizing and Improving Tag-Based Search in Folksonomies 113
relevant information for facilitating the resources retrieval and optimizing the time
expended during this process. Our paper is organized as follows: Section 2 presents a
quick overview about the main contributions attached to our search field; in Section 3
we will detail the design of our approach. After in Section 4 we move to the
experimental phase in order to measure the performance of our approach and discuss
the obtained results. Conclusion and future works are discussed in Section 5.
2 Related Work
In this section, we will put the point on the famous works which try to reduce the
tags’ ambiguity problem and especially those aimed to extract the semantic links
between folksonomy’s terms using ontologies. Mika [7] proposed to extend the
traditional bipartite model of ontologies to a tripartite one: that of folksonomies. In
another work, Gruber [5] argued that there is no contrast between ontologies and
folksonomies, and therefore recommended to build an "ontology of folksonomy".
According to Gruber, the problem of the lack of semantic links between terms in
folksonomies can be easily resolved by representing folksonomies by ontologies.
Specia and Motta [9] in their turn have preferred the use of ontologies to extract the
semantics of tags. Their approach consists in building tags clusters, and then trying to
identify the possible relationships between tags in each cluster. The niceTag project of
Limpens et al. [6] is focused on this same principle: the use of ontologies to extract
semantic links existing between tags in a system. In addition, this project has
introduced the idea of exploiting interactions between users and the system. Pan et al.
[8] aimed at reducing the problem of ambiguity in tagging. They proposed to extend
the search of tags in a folksonomy by using ontologies. They defended this principle
of extension of the search in order to avoid bothering the users with the rigidity of
ontologies. Beldjoudi et al. [1] proposed a technique specially designed to show the
social interactions’ usefulness in folksonomies for reducing tags’ ambiguity problem.
In another contribution the one of Beldjoudi el al. [2], the authors propose a method to
analyze user profiles according to their tags in order to personalize the
recommendation of resources. To sum up, most of the works relative to folksonomies
aim to bring together ontologies and folksonomies as a solution to the tags’ ambiguity
problem and that of the lack of semantic links between tags. In this context, we started
our trial to improve a little this technology and give a new view concerning the
combination between folksonomies and ontologies.
3 Semantic Social Folksonomy with Ontology (SSFO)
Our aim in this contribution is to introduce both the semantics and the social aspects
in folksonomies in order to let any user in the system retrieving relevant web
resources close to his preferences. In this paper, we aim to show how we can produce
a technique for helping any ontology already used for representing a folksonomy to
overcome the problem of tags’ ambiguity automatically without the need of an expert
who must control and organize links between terms. In addition we want show how

114 S. Beldjoudi, H. Seridi-Bouchelaghem, and C. Faron-Zucker
we can enrich our folksonomy (without human intervention) with relevant data in
order to help optimizing the time of search and enormously reduce the problems of
spelling variations and the lack of semantics within folksonomies focusing on the
rules-based systems.
3.1 Formal Description
Formally, a folksonomy is a tuple F = <U, T, R, A> where U, T and R represent
respectively the set of users, tags and resources, and A represents the relationship
between the three preceding elements i.e. A U x T x R. Because this approach is
intended to present a technique that can help any folksonomy represented by an
ontology to overcome the problems of tags’ ambiguity and spelling variations based
on the preferences and the interests of each user, and also enrich automatically the
system by new relevant data, we suggest here to represent our folksonomy with a
simple ontology defined by primitives relations such as "tagged by" and "used by"…
etc.
3.2 Resolving Tags’ Ambiguity in Folksonomies
Our technique to overcome the problem of tags’ ambiguity is not based on ontologies.
The idea is to study the profile of each member in the system and then compare the
preferences of this one with other users in order to extract those who are similar to
him.
It should be noted that: To make the system flexible, we propose to make it interact
with the user to accept or reject the retrieved resources. And to avoid the "cold start"
problem which is generally occur from a lack of the required data by the system in
order to make an excellent recommendation; it’s proposed to measure the similarity
between resources when the users are not similar. So we can summarize our
methodology as follow:
Similarities between Users. To calculate this similarity we suggest to use a measure
that allows representing each user by a vector v
i
designates a series of binary numbers
defined the set of his tags or his resources. Thus, to calculate the similarity between
two users, for example U
1
and U
2
, this measure proposes to calculate the cosines of
the angle between their associated vectors v
1
and v
2
as shown in the formula (1):
󰇛
,
󰇜
.
(1)
Similarities between Resources. When the users are not similar we suggest
measuring the degree of similarity between resources in order to avoid "cold start"
problem which is generally resulted from a lack of the data required by the system in
order to make an excellent recommendation.
Recommendation Levels. We propose here assigning to each resource recommended
by the system a factor that indicates the percentage of its recommendation. To achieve
this classification, we propose to calculate the ratio between the number of resources

Personalizing and Improving Tag-Based Search in Folksonomies 115
used by the user himself (i.e. the one who does the search) and the number of the
resources shared between him and the other users. Above a threshold fixed in [0..1],
we qualify the resource as highly recommended; under this threshold, it is simply
recommended or weakly recommended if the similarity is close to zero.
3.3 Rules-Based Systems in Folksonomies
The purpose of using rules-based systems can be summarized as follow: 1) Avoid the
existence of an expert who must control and organize links between terms. This let us
say that our technique is dynamic and automatic. 2) Optimize and reduce the time
required for searching relevant resources for each user by avoiding the recalculation
of similarities every time. And 3) enrich the folksonomy by a relevant fact which can
help improving the process of search and reducing the problem of spelling variations.
In our approach the folksonomies’ enrichment is realized by two categories of data as
follows:
1. Enrich our fact base by facts extracted from the similarities’ calculations that
have been made during the step 3.2; and which say that: such resource is similar to
such resource. For example; if we have already found that a resource R1 is similar to
another resource R2, then we can add the following fact: is-similar-to (R1, R2) which
express that "R1 is similar to R2". With this method our system does not recalculate
the similarity between the users every time when an actor want to search relevant
resources, but it will optimize this time and also the memory space that can be lost in
each calculation because with this process; before our system begin the calculation of
similarity between users or between resources it will firstly see in the fact base if there
are resources similar to those already proposed to this user.
2. The second kind of facts has the following form: "A resource R
Z
can have as
tags the tag T
Y
" or can-tagged-by (R
Z
, T
Y
). The advantage of such fact is twofold: a)
Reduce the problem of tags’ ambiguity (because the similarity between resources
became more highly). b) Reduce the problem of spelling variations. We can explain
this second point (b) by the following example: "cat" and "chat" means both the same
concept (animal) in English and in French, but when a user searches resources
annotated by the tag "cat", the system will not offer him those tagged by the word
"chat" because it can’t understand that the tag "cat" is equivalent to the tag "chat". In
others words, supposing that the user U
X
tagged a resource R
1
by the tag cat and U
W
is the user who tagged the resource R
2
by the tag chat. Noting that; the two resources
R
1
and R
2
are already considered as similar according to the similarities’ calculations
that have been made before. Now if the user U
X
wants search resources concerning
the animal "cat" by the tag cat, the resource R
2
will not be given to him. In order to
overcome this problem our approach proposes to add the following facts: can-tagged-
by (R
1
, chat), can-tagged-by (R
2
, cat). And now any user can benefit from the
resources of the other and so we have overcame the problem of spelling variation in
folksonomies.

Citations
More filters
Book ChapterDOI
06 Jun 2016
TL;DR: A social personalized ranking function is proposed; this function leverages the social aspect of folksonomy and events detection to estimate the relevance of given resources to a tag-based query issued by learners.
Abstract: Recently, the social web has recognized a real attention by E-learning community. This collaborative space gave students new opportunities to share their contents and receive immediate feedback from other networkers. For instance, in folksonomies, learners are able to tag useful resources within a highly visible space, which allow sharing ideas that gives a basis for discussion, and thus other students can benefit from those resources. Actually, social environments offer a unique opportunity to personalize search spaces. The objective of this work is to achieve this opportunity and thus personalize tag-based search in E-learning folksonomy by extract implicitly the semantics of learners’ tags. In this context, a social personalized ranking function is proposed; this function leverages the social aspect of folksonomy and events detection to estimate the relevance of given resources to a tag-based query issued by learners.

3 citations

Proceedings ArticleDOI
10 Nov 2012
TL;DR: The aim of this contribution is to provide another view about the semantics missed in the social web technologies and show how to use the methods of Data mining to extract the meaning of tags in folksonomies.
Abstract: In these recent years, the social semantic web has recognized a real attention by the majority of researchers in the world We propose in this paper an original approach concerning a powerful technology which has recognized a great success in the social web area, we talk about folksonomies The aim of our contribution is to provide another view about the semantics missed in the social web technologies and show how we can use the methods of Data mining to extract the meaning of tags in folksonomies Especially we will present how we can analyze user profiles according to their tags in order to personalize the recommendation of pertinent resources and surmount the lack of semantic links between tags for improving the quality of results returned by the current folksonomies

Cites background from "Personalizing and improving tag-bas..."

  • ...In this work the authors argue that the automatic sharing of resources strengthens social links among actors and exploit this idea to enrich user profiles by increasing the weights associated to web resources according to social relations....

    [...]

References
More filters
BookDOI
01 Jan 2004
TL;DR: DODDLE-R, a support environment for user-centered ontology development, consists of two main parts: pre-processing part and quality improvement part, which generates a prototype ontology semi-automatically and supports the refinement of it interactively.
Abstract: In order to realize the on-the-fly ontology construction for the Semantic Web, this paper proposes DODDLE-R, a support environment for user-centered ontology development. It consists of two main parts: pre-processing part and quality improvement part. Pre-processing part generates a prototype ontology semi-automatically, and quality improvement part supports the refinement of it interactively. As we believe that careful construction of ontologies from preliminary phase is more efficient than attempting generate ontologies full-automatically (it may cause too many modification by hand), quality improvement part plays significant role in DODDLE-R. Through interactive support for improving the quality of prototype ontology, OWL-Lite level ontology, which consists of taxonomic relationships (class sub class relationship) and non-taxonomic relationships (defined as property), is constructed effi-

2,006 citations

Book ChapterDOI
06 Nov 2005
TL;DR: This work extends the traditional bipartite model of ontologies with the social dimension, leading to a tripartite modelof actors, concepts and instances, and demonstrates the application of this representation by showing how community-based semantics emerges from this model through a process of graph transformation.
Abstract: In our work we extend the traditional bipartite model of ontologies with the social dimension, leading to a tripartite model of actors, concepts and instances We demonstrate the application of this representation by showing how community-based semantics emerges from this model through a process of graph transformation We illustrate ontology emergence by two case studies, an analysis of a large scale folksonomy system and a novel method for the extraction of community-based ontologies from Web pages

815 citations

Book
01 Dec 2005
TL;DR: Semantic Acceleration Helping Realize the Semantic Web Vision or "The Practical Web", research/Academic track.

546 citations

BookDOI
01 Jan 2005
TL;DR: In this paper, the authors present an approach for using the Semantic Web for e-Science: Inspiration, Incubation, Irritation, and Semantic Acceleration Helping Realize the semantic Web Vision or "The Practical Web".
Abstract: Invited Paper.- Using the Semantic Web for e-Science: Inspiration, Incubation, Irritation.- Semantic Acceleration Helping Realize the Semantic Web Vision or "The Practical Web".- Semantic Web Public Policy Challenges: Privacy, Provenance, Property and Personhood.- Research/Academic Track.- Constructing Complex Semantic Mappings Between XML Data and Ontologies.- Stable Model Theory for Extended RDF Ontologies.- Towards a Formal Verification of OWL-S Process Models.- Web Service Composition with Volatile Information.- A Large Scale Taxonomy Mapping Evaluation.- RDF Entailment as a Graph Homomorphism.- RitroveRAI: A Web Application for Semantic Indexing and Hyperlinking of Multimedia News.- Querying Ontologies: A Controlled English Interface for End-Users.- Semantic Browsing of Digital Collections.- Decentralized Case-Based Reasoning for the Semantic Web.- Finding and Ranking Knowledge on the Semantic Web.- Choreography in IRS-III - Coping with Heterogeneous Interaction Patterns in Web Services.- Bootstrapping Ontology Alignment Methods with APFEL.- A Strategy for Automated Meaning Negotiation in Distributed Information Retrieval.- On Applying the AGM Theory to DLs and OWL.- A General Diagnosis Method for Ontologies.- Graph-Based Inferences in a Semantic Web Server for the Cartography of Competencies in a Telecom Valley.- Ontology Design Patterns for Semantic Web Content.- Guidelines for Benchmarking the Performance of Ontology Management APIs.- Semantically Rich Recommendations in Social Networks for Sharing, Exchanging and Ranking Semantic Context.- On Partial Encryption of RDF-Graphs.- Seven Bottlenecks to Workflow Reuse and Repurposing.- On Logical Consequence for Collections of OWL Documents.- A Framework for Handling Inconsistency in Changing Ontologies.- Preferential Reasoning on a Web of Trust.- Resolution-Based Approximate Reasoning for OWL DL.- Reasoning with Multi-version Ontologies: A Temporal Logic Approach.- Piggy Bank: Experience the Semantic Web Inside Your Web Browser.- BRAHMS: A WorkBench RDF Store and High Performance Memory System for Semantic Association Discovery.- A Template-Based Markup Tool for Semantic Web Content.- Representing Web Service Policies in OWL-DL.- Information Modeling for End to End Composition of Semantic Web Services.- Searching Dynamic Communities with Personal Indexes.- RUL: A Declarative Update Language for RDF.- Ontologies Are Us: A Unified Model of Social Networks and Semantics.- OMEN: A Probabilistic Ontology Mapping Tool.- On the Properties of Metamodeling in OWL.- A Bayesian Network Approach to Ontology Mapping.- Ontology Change Detection Using a Version Log.- RelExt: A Tool for Relation Extraction from Text in Ontology Extension.- Containment and Minimization of RDF/S Query Patterns.- A String Metric for Ontology Alignment.- An Ontological Framework for Dynamic Coordination.- Introducing Autonomic Behaviour in Semantic Web Agents.- Combining RDF and Part of OWL with Rules: Semantics, Decidability, Complexity.- Benchmarking Database Representations of RDF/S Stores.- Towards Imaging Large-Scale Ontologies for Quick Understanding and Analysis.- Automatic Evaluation of Ontologies (AEON).- A Method to Combine Linguistic Ontology-Mapping Techniques.- Debugging OWL-DL Ontologies: A Heuristic Approach.- Rapid Benchmarking for Semantic Web Knowledge Base Systems.- Using Triples for Implementation: The Triple20 Ontology-Manipulation Tool.- A Little Semantic Web Goes a Long Way in Biology.- Provenance-Based Validation of E-Science Experiments.- Industrial Track.- Semantic Service Integration for Water Resource Management.- Towards a Killer App for the Semantic Web.- Enterprise Architecture Reference Modeling in OWL/RDF.- MediaCaddy - Semantic Web Based On-Demand Content Navigation System for Entertainment.- LKMS - A Legal Knowledge Management System Exploiting Semantic Web Technologies.- Definitions Management: A Semantics-Based Approach for Clinical Documentation in Healthcare Delivery.- Ubiquitous Service Finder Discovery of Services Semantically Derived from Metadata in Ubiquitous Computing.- Ontological Approach to Generating Personalized User Interfaces for Web Services.- On Identifying Knowledge Processing Requirements.- An Application of Semantic Web Technologies to Situation Awareness.- Task Knowledge Based Retrieval for Service Relevant to Mobile User's Activity.- Supporting Rule System Interoperability on the Semantic Web with SWRL.- Automated Business-to-Business Integration of a Logistics Supply Chain Using Semantic Web Services Technology.- A Semantic Search Engine for the International Relation Sector.- Gnowsis Adapter Framework: Treating Structured Data Sources as Virtual RDF Graphs.- Do Not Use This Gear with a Switching Lever! Automotive Industry Experience with Semantic Guides.- The Concept Object Web for Knowledge Management.- Semantic Web Challenge.- The Personal Publication Reader.- DynamicView: Distribution, Evolution and Visualization of Research Areas in Computer Science.- Oyster - Sharing and Re-using Ontologies in a Peer-to-Peer Community.- The FungalWeb Ontology: Semantic Web Challenges in Bioinformatics and Genomics.- CONFOTO: A Semantic Browsing and Annotation Service for Conference Photos.

530 citations

Frequently Asked Questions (1)
Q1. What contributions have the authors mentioned in the paper "Personalizing and improving tag-based search in folksonomies" ?

The authors present in this paper an original approach concerning a technology that has recognized a great popularity in these recent years, they talk about folksonomies. Their aim in this contribution is propose new technique for the Social Semantic Web technologies in order to see how the authors can overcome the problem of tags ’ ambiguity automatically in folksonomies even when they choose representing these latter with ontologies. The authors ’ ll also illustrate how they can enrich any folksonomy by a set of pertinent data to improve and facilitate the resources ’ retrieval in these systems ; all this with tackling another problem, they speak about spelling variations.