2010 IEEE Fourth International Conference on Semantic Computing 2010 IEEE International Conference on Semantic Computing

A Semantic Web Personalizing Technique

The case of bursts in web visits
Dimitris Antoniou, Mersini Paschou, Efrosini Sourla, Athanasios Tsakalidis Computer Engineering and Informatics Department University of Patras Patras 26504, Greece {antonid, paschou, sourla, tsak}@ceid.upatras.gr

Abstract The explosive growth in the size and use of the World Wide Web continuously creates new great challenges and needs. The need for predicting the users' preferences in order to expedite and improve the browsing though a site can be achieved through personalizing of the websites. A personalization mechanism is based on explicit preference declarations by the user and on an iterative process of monitoring the user navigation, collecting its requests of ontological objects and storing them in its profile in order to deliver personalized content. The problem that we address is the case where few web pages become very popular for short periods of time and are accessed very frequently in a limited temporal space. Our aim is to deal with these bursts of visits and suggest these highly accessed pages to the future users that have common interests. Hence, in this paper, we propose a new web personalization technique, based on advanced data structures. The data structures that are used are the Splay tree (1) and Binary heaps (2). We describe the architecture of the technique, analyze the time and space complexity and prove its performance. In addition, we compare both theoretically and experimentally the proposed technique to another approach to verify its efficiency. Our solution achieves O(P2) space complexity and runs in klogP time, where k is the number of pages and P the number of ontologies of Web pages. Keywords-component; personlization, ontologies, adaptive data structure

is not always obvious. Enriching the Web application with personalized recommendations provides alternative paths to published data, and increments the possibilities for the user to find the contents he is interested in. However, the effectiveness of personalization is based on the quality of the user profile and of the relations among the content objects. Modeling the published data and the user profile with ontologies allows to express more effectively the user interests and the relations between the pieces of information, by leveraging the advanced features of Semantic Web technologies. Such semantic relations may be exploited for more accurate personalization results. A personalization mechanism is based on explicit preference declarations by the user and on an iterative process of monitoring the user navigation, collecting its requests of ontological objects and storing them in its profile in order to deliver personalized content. We aim at storing relational data between ontological objects concerning their popularity and user requests on ontological objects related to its navigation. Recommendation and personalization algorithms aim at suggesting web pages to users based on their current visit and past users' navigational patterns. The problem that we address is the case where few web pages become very popular for short periods of time and are accessed very frequently in a limited temporal space. Our aim is to deal with these bursts of visits and suggest these highly accessed pages to the future users that have common interests. Hence, in this paper, we propose a new web personalization technique, based on advanced data structures. The data structures that are used are the Splay tree (1) and Binary heaps (2). We describe the architecture of the technique, analyze the time and space complexity and prove its performance. In addition, we compare both theoretically and experimentally the proposed technique to another approach to verify its efficiency. Our solution achieves O(P2) space complexity and runs in klogP time, where k is the number of pages and P the number of ontologies of Web pages. II. PREVIOUS WORK

I.

INTRODUCTION

The Semantic Web leverages the knowledge integration on the Web to new levels. Despite the efforts put into the technical and research issues, there are few applications actually deploying and evaluating semantic web with real users. Semantic web can only deliver if it is driven by user needs, context or profiles to seamlessly integrate the knowledge on the web to really provide desirable content. Context and customization are some of the key factors for accurate, effective relevant information access in Internet digital libraries and in general in the Semantic Web. Within traditional Web applications, the user navigation follows the predefined hypertext structure. Therefore, finding contents requires the user understanding of the Web site outline, which

Website personalization has become an important issue because of the popularity of e-commerce applications [1,7,9].
524 530

978-0-7695-4154-9/10 $26.00 2010 IEEE DOI 10.1109/ICSC.2010.49

Several methods for website personalization have been proposed [1,3,4,6]. The goal of a personalized website is to take advantage of the knowledge obtained from the analysis of the user's navigational behavior in combination with other information collected, such as the users location, previous navigation patterns, and items purchased [16,5,6,7,9]. Another very important issue is the structure of the website and the statistical study of the links and pages that it consists of. UPR is a PageRank-style algorithm which combines usage data and link analysis techniques for assigning probabilities to Web pages based on their importance in the Web site's navigational graph [15]. Another widely used web personalization technique is the web usage data mining personalization [1]. For example, a classification algorithm for web personalization based on web usage mining techniques has been proposed. The algorithm takes into account both static information, by means of classical clustering techniques, and dynamic user behavior, thus proposing a novel and effective re-classification algorithm [17]. Computational Intelligent combinations have been also applied in the context of Web personalization, by providing different examples of intelligent systems which have been designed to provide Web users with the information they search, without expecting them to ask for it explicitly [19]. On the other hand, a novel online recommender system builds profiling models and offers suggestions without the user taking the lead [18]. In general, personalization recently has been employed in several other fields as well. In advertising, a new online advertisement targeting technique is proposed [28], which adapts and utilizes several powerful and well tested information retrieval and lexical techniques to develop an estimate of a user's affinity for particular products and services based on an analysis of a user's web surfing behavior. Moreover, a design research approach combining behavior and engineering techniques is proposed to better support user modeling in personalized mobile advertising applications [29]. Even in Downloading mobile games via mobile device, a personalized mobile game recommendation system is proposed, which takes under consideration the time-of-day and time-of-week is used to provide a more personalized experience [31]. As far as bursts as an algorithmic issue are concerned, several papers have been presented as well. A new algorithmic framework for elastic burst detection has been introduced: a family of data structures that generalizes the Shifted Binary Tree, and a heuristic search algorithm to find an efficient structure given the input [27]. In addition, semantic personalization has been enhanced into digital libraries and web portals. Semantic Browsing provides contextualized dynamically generated Web content customizing the knowledge to better meet user expectations. For example, the real-world medical digital library, the National electronic Library of Infection (NeLI, www.neli.org.uk) [32], enriched with an infection domain ontology enables new semantic services to be developed qualitatively. In this work, group profiling was used to customize semantic browsing by integrating distributed

knowledge sources. The service is evaluated by web server logs analysis, dynamically enhancing the profiles and by qualitative feedback from real users of the NeLI portal. The Internet consists of web sites that employ different kinds of structures as the backbone of their build-up. However, users are browsing the web according to its content, regardless of the structure. In [33] the possibilities of applying ontologies in exploring the web sites structures and usage for producing viewing recommendations for the visitors are discussed. A special log system for capturing access data is introduced as well as techniques applied for data mining. Ontology of user profiles is constructed by exploiting the user locality model. Furthermore, enriching Web applications with personalized data is of major interest for facilitating the user access to the published contents, and therefore, for guaranteeing successful user navigation. In [34] a conceptual model for extracting personalized recommendations based on user profiling, ontological domain models, and semantic reasoning is proposed. The approach offers a high-level representation of the designed application based on a domain-specific metamodel for Web applications called WebML. Integrating usage data with content, structure or user profile data enhances the results of the personalization process. In [35], SEWeP is presented, a system that makes use of both the usage logs and the semantics of a Web sites content in order to personalize it. Web content is semantically annotated using a conceptual hierarchy (taxonomy). C-logs is introduced, an extended form of Web usage logs that encapsulates knowledge derived from the link semantics. C-logs are used as input to the Web usage mining process, resulting in a broader yet semantically focused set of recommendations. The challenge of the Semantic Web Mining technologies in the e-Learning domain can relate to the provision of personalized experiences for the users. Particularly, these applications can take into consideration the individual needs and requirements of learners. In [36], a framework for personalized e-Learning based on aggregate usage profiles and a domain ontology is proposed. The writers distinguished two stages in the whole process, one of offline tasks that includes data preparation, ontology creation and usage mining and one of online tasks that concerns the production of recommendations. Web usage mining has been used effectively as an approach to automatic personalization and as a way to overcome deficiencies of traditional approaches such as collaborative filtering. Despite their success, such systems, as in more traditional ones, do not take into account the semantic knowledge about the underlying domain. Without such semantic knowledge, personalization systems cannot recommend different types of complex objects based in their underlying properties and attributes. Nor can these systems possess the ability to automatically explain or reason about the user models or user recommendations. The integration of semantic knowledge is, in fact, the primary challenge for the next generation of personalization systems. In [37] an overview of approaches for incorporating semantic knowledge into Web usage mining and personalization processes is provided. In particular the issues and requirements for successful integration of semantic knowledge from different sources, such as the

531 525

content and the structure of Web sites for personalization are discussed. Finally, a general framework for fully integrating domain ontologies with Web Usage Mining and Personalization processes at different stages are presented, including the preprocessing and pattern discovery phases, as well as in the final stage where the discovered patterns are used for personalization.

accessed very frequently in a limited temporal space. Such patterns have been also observed in various Internet applications in a number of studies [10]. In bursty web search pattern cases, the user attempts to find specific results that belong to limited ontologies of interest within a short time period. As a consequence an efficient retrieval and storing mechanism is needed to keep the users personalized ontologies and frequent results. We have a set of ontologies of WebPages and a number of random visits are executed to all WebPages by users. We define a set of WebPages to be preferred by the user when these Webpages are the most highly visited by him/her based on the visits recorded for a certain time interval. In particular, we count for each WebPage, how many accesses have taken place since the last time it was visited. If this number is sufficient to denote this WebPage as preferred and the time interval during which the accesses are performed is satisfying, then this access pattern is considered as a burst of visits. Due to the overload of Webpages, dealing with ontologies offer more advantages, since it helps portioning the problem. With the constant increase of the number of web pages, it has become very difficult for a user to locate the desired information in the web or even in a website. For the ease of the user many websites could organize their Webpages into ontologies in order to aid the search of a WebPage by assigning it to a ontology. The same concept is adopted even by search engines. Hence, our goal is to take advantage of the organization of the WebPages into ontologies and use it in order to deal with the burst of visits to a certain ontology of WebPages. For example, let us suppose that a user usually visits the a certain ontology of an e-shop web site and due to a web commercial he visits the ontology of video and sound. At this point, a web personalization technique that deals with bursty visits, would suggest to the user WebPages and ontologies of WebPages that users that prefer the ontology of video and sound visit as well. IV. WEBPAGES AND ONTOLOGIES

III.

PERSONALIZATION AND BURSTS

Personalization can be defined as the design, management and delivery of content based on known, observed and predictive information. Personalization techniques match an individual, his/her preferences and Web page click stream habits with tailored content based on a user profile. In todays world of information overload many similar technologies are used as a way to filter and organize the data most important to them. Correctly executed, personalization of the visitors experience makes his time on a site, or in an application, more productive and engaging. Personalization can also be valuable for an organization, a portal or an e-store, because it drives desired business results such as increasing visitor response or promoting customer retention. In this work we try to enhance the case of burst of visits in the personalizing of web sites. Many aspects of everyday life are described by events [27]. An unexpectedly large number of events occurring within a certain time period is called a burst, suggesting unusual actions or processes. Bursts may occur in many everyday situations from economics to natural phenomena, such as trading stocks and falling stars. Depending on the importance of the phenomenon or the process observed, efficiently detecting bursts is critical. By definition, a burst depends on the temporal region we are focusing on, that is the window size. Bursts occur in a Websites traffic as well and affect its functionality in many aspects as the one it follows. As more and more commercial enterprises go online, it is vital to make their websites attractive to customers. One way to attract website traffic is online advertising on search engines. In this case, besides the search results, an ad is placed in the search engines webpage. If the visitor clicks the ad, the advertiser has to pay a fee to the search engine. A problem that has arisen with pay-per-click is click fraud. Someone can use an automated script or program to simulate multiple clicks by a browser on an ad. Of course, the number of clicks has to be large enough in order to gain a considerable amount of money. Therefore, a burst of clicks may be an indication of a click fraud. In this paper, we deal with the case of a burst of visits to a Webpage and how someone can gain knowledge from this fact and aid the personalization of the web. A pattern of visits or accesses is bursty when they occur with high intensity over a limited period of time. In particular, in bursty cases, a few web pages become very popular for short periods of time and are

Before describing our personalization algorithm, we have to explain how the web pages can be assigned to ontologies. Ontology in computer science is the representation of entities, ideas, and events, along with their properties and relations, according to a system of ontologies. We will use the ORGAN Web log analysis tool that offers an integrated solution of building and performing analysis, taking into consideration both the site content semantics and the Web site page visits. Information about the user preferences in relation to the Web site topics may be extracted, which will constitute ORGAN as a valuable application during the site administrators decisionmaking about the Web site reorganization. Hence, before applying our personalization algorithm on a site, we use the ORGAN tool to assign the Web pages to proper ontologies. V. THE PROBLEM

We describe the problem as follows: We have a set of P ontologies of WebPages and N users. Each Webpage belongs to certain ontology and each users profile is kept in a splay

532 526

tree. As profile, we define the logfile of the WebPages that the user has visited so far. In the splay tree, we store the ontologies of the Web Pages. According to the splay trees properties, the item that was last visited is brought to the root of the tree. In our case, we modify the tree, so that the most frequently accessed item is the one splayed to the root. In fact, we intend to splay a ontology when we observe a burst of visits to it. After a ontology is splayed to the root due to a burst of hits, no other reconstruction or splaying is needed. Therefore, the ontologies that appear in the upper levels of the splay tree of each user are the ones preferred by the user.

work is the binary heap. Each ontology keeps a binary heap with the rest of the ontologies and their popularities. According to the priority queues properties, in the root we keep the minimum key and hence we can access it in O(1) time. We store the number of popularity of each ontology with negative sign, in order to keep the maximum in the root of the priority queue. Every time we observe a burst of visits to a ontology A, we increase by one the counter of the popularity of this ontology in all the priority queues of the ontologies found in the top levels of the splay tree of the particular user. In addition, we increase the counters of these ontologies in the priority queue of the ontology A. Hence, we are able to extract the most popular ontology to people that visited ontology A in constant time. Depending on whether in the splay trees we store ontologies of WebPages or WebPages, there can be a different approach. In the case of WebPages, to ensure that, at least a big portion of pages of a ontology is splayed to the root, we can follow the following technique. We suppose that a page x is accessed times k, which are enough to denote this page as the most frequently accessed page. Then, as long as the father of this node belongs to the same ontology, we traverse the tree bottom-up. The upper threshold of the number of the levels that we climb depends on the number of ontologies and WebPages. Let z be the last predecessor of x that belongs to the same ontology with x. Then, we splay all the nodes belonging to its sub tree to the root. In addition, we want to distinguish/mark the ontologies most visited by each user. The ontologies closer to the root of the tree, are those preferred by the user. Hence, the only thing left to decide is the depth of the tree that will represent the bound, above which all ontologies are concerned preferred.

1.

While (A webpage of ontology W is accessed by a user) /*Retrieve information from the Webpages logfile*/

2. 3.

if (this access has created a burst of visits to the ontology W) then

/* Rearrange the splay tree of the user, so that the ontology with the latest bursty access pattern is brought to the root */ 4. Splay the ontology W to the root of the users splay tree

Figure 1 Architecture of the Personalization Algorithm

/* Update the priority queues of the ontologies in order to promote the popular ontologies of the user */ 5. Define the set of ontologies ,TOP, that already exist in the top levels of the splay tree 6. Increase the counter of W in all the priority queues of the ontologies that belong to TOP 7. Increase the counter of the ontologies that belong to TOP in the priority queue of W

Since we keep a profile for each user, we wish to build a data structure that will keep the most popular ontologies that most users that visited A for example, prefer to visit as well. Therefore, for each ontology we build a priority queue. The priority queue that we have chosen for the purposes of this

533 527

8. 9.

Return as suggested the root of the priority queue of W endif

10. else 11. continue

complexity and runs in klogP time, where k is the number of pages and P the number of ontologies. Future steps include the evolution of the algorithms to take into account additional implicit user feedback of the final products chosen and not only the e-shops and services. This is particularly efficient case for e-businesses implementations based on lightweight RESTful mobile Web Services. REFERENCES
[1] 1. Cho, Y., Kim, J., and Kim, S. (2002) A personalized recommender system based on web usage mining and 20th Computer Science Seminar SF2-T2-4 decision tree induction. Expert Systems with Applications. 23, 329-342. 2. Dayal, U., Hanson, E., and Widom, J. (1994) Active Database Systems. In: Modern Database Systems: The Object Model, Interoperability, and Beyond. Kim, W., ed. Addison-Wesley, Reading, MA. 3. Fong, J., Hughes, J., and Zhu, J. (2000) Online Web Mining Transactions Association Rules using Frame Metadata Model. IEEE. 121-129. 4. Garrigs, I., Gmez, J., Cachero, C. (2003) Modelling Dynamic Personalization in Web Applications. ICWE 2003. 472-475. 5. Hu, S. (2002) Helping Online Customers Decide through Web Personalization. IEEE. 17, 34-43. 6. Kiyomitsu, H., Takeuchi, A., Tanaka, K. (2001) Web Reconfiguration by Spatio-Temporal Page Personalization Rules Based on Access Histories. In: Proceedings of the Symposium on Applications and the Internet (SAINT 2001). IEEE Press. 75-82. 7. VanderMeer, D., Dutta, K., and Datta, A. (2000) Enabling Scalable Online Personalization on the Web. In: Proceedings of EC00, October 17-20, 2000, Minneapolis, Minnesota. 185-196. 8. Wu, D., Im, I., Tremaine, M., Instone, K., Turoff, M. (2002) A Framework for Classifying Personalization Scheme Used on eCommerce Websites. IEEE Proceedings of the 36th Hawaii International Conference on System Sciences (HICSS03). 9. Yu, P. (1999) Data Mining and Personalization Technologies. In: Proceedings of the International Conference on Database Systems for Advanced Applications (DASFAA99). 1-22.20th Computer Science Seminar SF2-T2-5 10. Zhou, B., Hui, S.C., and Chang, K., An Intelligent Recommender System Using Sequential Web Access Patterns, IEEE Conference on Cybernetics and Intelligent Systems. IEEE, Singapore, pages 393398, 2004 11. Ergn, F., Mittra, S., Cenk Sahinalp, S., Sharp, J., Sinha, R.K.: A Dynamic Lookup Scheme for Bursty Access Patterns. INFOCOM 2001, pp. 1444-1453 12. Lin, S., and McKeown, N., A simulation study of IP switching, in Proc. ACM SIGCOMM, 1997. 13. Jebari Chaker, Ounelli Habib, "Genre Categorization of Web Pages," Data Mining Workshops, International Conference on, pp. 455464, Seventh IEEE International Conference on Data Mining Workshops (ICDMW 2007), 2007. 14. D.D. Sleator and R.E. Tarjan. Self-adjusting binary search trees, Journal of the ACM, 32:652-686, 1985. 15. Magdalini Eirinaki, Michalis Vazirgiannis: Web site personalization based on link analysis and navigational patterns. ACM Trans. Internet Techn. 7(4): (2007) 16. Magdalini Eirinaki, Michalis Vazirgiannis: Web mining for web personalization. ACM Trans. Internet Techn. 3(1): 1-27 (2003) 17. Antonio Picariello, Carlo Sansone A web usage mining algorithm for web personalization Source Intelligent Decision Technologies archive Volume 2 , Issue 4 (October 2008) Pages: 219-230 Year of Publication:2008

VI.

ANALYSIS

A. Space requirements As far as the space complexity is concerned, the space, as expected, is mainly consumed by the two data structures. Splay trees: We need 1 splay tree for each user. In worst case, in each splay tree we store W WebPages. Therefore, taking under consideration the extra fields needed for each node of the splay tree, the space required is 5NW. Priority queues: For each ontology we keep a priority queue. Hence, for P ontologies, we need O(P2) space. B. Time requirements As far as time complexity is concerned, per access we need:

[2]

[3]

[4] [5] [6]

log(

 W ) , in order to splay nodes. That is at most  w

[7]

log(#pages). We need O(1) time to return the root from each priority queue. Hence, N O(1) time is required to suggest a ontology to N users. Finally, we need to update the priority queues. In other words, before suggesting the root of a ontologys priority queue, we have to increase the keys of the ontologies that we find in the users favorites if they are already in the priority queue of the splayed ontology. Finally, we have to increase the key of the ontology splayed, in all the priority queues of the top levels of the splay tree of the user. So, in total, klogP time is needed. VII. CONCLUSIONS AND FUTURE WORK Recommendation and personalization algorithms aim at suggesting web pages to users based on their current visit and past users' navigational patterns. In this paper, we propose a web personalization technique, based on advanced data structures. The main concept of this work is to deal with the case of a burst of visits to a WebPage by designing an algorithm that suggests to the visitors of a certain ontology of WebPages A, other ontologies of WebPages that previous visitors of A prefer to visit as well. The data structures that are used are the Splay tree (1) and Binary heaps (2). We describe the architecture of the technique and analyze the time and space complexity. Our solution achieves O(P2) space

[8]

[9]

[10]

[11]

[12] [13]

[14] [15]

[16] [17]

534 528

[18] 18. Ranieri Baraglia, Fabrizio Silvestri: Dynamic personalization of web sites without user intervention. Commun. ACM 50(2): 63-67 (2007) [19] 19. G. Castellano, A.M. Fanelli and M. A. Torsello (2008), Computational Intelligence Techniques for Web personalization. Web Intelligence and Agent Systems: an International Journal, VOL. 6, Number 3/2008, pages 253-272. Publisher: IOS Press, Amsterdam, The Netherlands. ISSN 1570-1263 [20] 20. Knuth, D.E. The Art of Computer Programming, Vol. 111 Sorting and Searching. Addison-Wesley, Reading, Mass., 1973. [21] 21. Atkinson, M.D., J.-R. Sack, N. Santoro, and T. Strothotte (1 October 1986). "Min-max heaps and generalized priority queues.". Programming techniques and Data structures. Comm. ACM, 29(10): 996-1000. [22] 22. Lin, S., and McKeown, N., A simulation study of IP switching, in Proc. ACM SIGCOMM, 1997. [23] 23. Zhou, B., Hui, S.C., and Chang, K., An Intelligent Recommender System Using Sequential Web Access Patterns, IEEE Conference on Cybernetics and Intelligent Systems. IEEE, Singapore, pages 393398, 2004 [24] 24. Ergn, F., Mittra, S., Cenk Sahinalp, S., Sharp, J., Sinha, R.K.: A Dynamic Lookup Scheme for Bursty Access Patterns. INFOCOM 2001, pp. 1444-1453 [25] 25. Wei Jiang, Chen Ding, Roland Cheng :Memory access analysis and optimization approaches on splay treesLCR '04: Proceedings of the 7th workshop on Workshop on languages, compilers, and run-time support for scalable systems [26] 26. T. Srinivasan, M. Nivedita and V. Mahadevan, Efficient Packet Classification using Splay Tree Models, IJCSNS, Vol 6, No. 5B, May 2005 [27] 27. Xin Zhang , Dennis Shasha, Better Burst Detection, Proceedings of the 22nd International Conference on Data Engineering, p.146, April 0307, 2006 [28] 28. Jason Deane , Praveen Pathak, Ontological analysis of web surf history to maximize the click-through probability of web advertisements, Decision Support Systems, v.47 n.4, p.364-373, November, 2009 [29] 29. David Jingjun Xu , Stephen Shaoyi Liao , Qiudan Li, Combining empirical experimentation and modeling techniques: A design research approach for personalized mobile advertising applications, Decision Support Systems, v.44 n.3, p.710-724, February, 2008 [30] 30. Nan Li , Desheng Dash Wu, Using text mining and sentiment analysis for online forums hotspot detection and forecast, Decision Support Systems, v.48 n.2, p.354-368, January, 2010 [31] 31. Worapat Paireekreng, Kowit Rapeepisarn, Kok Wai Wong: TimeBased Personalised Mobile Game Downloading. T. Edutainment 2: 5969 (2009) [32] Patty Kostkova, Gayo Diallo, Gawesh Jawaheer: User Profiling for Semantic Browsing in Medical Digital Libraries. ESWC 2008: 827-831 [33] Tarmo Robal, Ahto Kalja: Applying User Profile Ontology for Mining Web Site Adaptation Recommendations. ADBIS Research Communications 2007 [34] Christina Tziviskou, Marco Brambilla: Semantic Personalization of Web Portal Contents, WWW 2007: 1245-1246 [35] Magdalini Eirinaki, Michalis Vazirgiannis, Iraklis Varlamis: SEWeP: using site semantics and a taxonomy to enhance the Web personalization process. KDD 2003: 99-108 [36] Penelope Markellou, Ioanna Mousourouli, Sirmakessis Spiros, Athanasios Tsakalidis, Using Semantic Web Mining Technologies For Personalized E-Learning Experiences, in the Proceedings of the

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

IASTED International Conference on Web based Education (WBE 2005), pp 522-527, February 21-23, 2005, Grindelwald, Switzerland. Dai H., & B. Mobasher, "Integrating semantic knowledge with web usage mining for personalization", Web Mining: Applications and Techniques, A. Scime (Ed.), Hershey: Idea Group Publishing, 2004, 276-306. Ch. Makris , Y. Panagis, E. Sakkopoulos, and A. Tsakalidis,(2007), Category ranking for personalized search, in the Data and Knowledge Engineering Journal (DKE), Elsevier Science, Volume 60, Issue 1 , January 2007, Pages 109-125 E. Sakkopoulos, D. Kanellopoulos and A. Tsakalidis (2006), Semantic mining and web service discovery techniques for media resources management, in Int. J. Metadata, Semantics and Ontologies, Vol. 1, No. 1, pp.6675 J. Garofalakis, Y. Panagis, E. Sakkopoulos, and A. Tsakalidis (2006), Optimization mechanism for Web Search Results using Topic Knowledge, in the International Journal of Knowledge and Learning, Vol. 2 (1&2), pp. 140-153 E. Christopoulou, J. Garofalakis, Ch. Makris, Y. Panagis, E. Sakkopoulos, A. Tsakalidis (2003), Techniques and Metrics for Improving Website Structure, Journal of Web Engineering, Vol.2 No.1&2 pp. 090-104 Dimitris Antoniou, John D. Garofalakis, Christos Makris, Yannis Panagis, Evangelos Sakkopoulos: Context-similarity based hotlinks assignment: Model, metrics and algorithm. Data Knowl. Eng. 69(4): 357-370 (2010) Makris, Ch., Panagis, Y., Plegas, Y., Sakkopoulos, E., An integrated Web System to facilitate personalized web searching algorithms, ACM 23rd Annual ACM Symposium on Applied Computing (ACM SAC 2008), Track on Web Technologies, Cear, Brazil, March 16 - 20, 2008, pp. 2397-2402 John Garofalakis, John Giannakou, Theodoula Giannakoudi, Evangelos Sakkopoulos: Promo: Novel metrics that combine hits and accessibility criteria in order to promote important content and links, Euro American Conference on Telematics and Information Systems, ACM, 2007, ISBN # 978-1-59593-598-4 John Garofalakis, Theodoula Giannakoudi, Evangelos Sakkopoulos , An integrated technique for Web Site Usage Semantic Analysis: The ORGAN System, International Workshop Logging Traces of Web Activity: The Mechanics of Data Collection in conjunction with the WWW 2006, in the Workshop Proceedings, Edinburg, Scotland, May 23, 2006 J. Garofalakis, Th. Matsoukas, Y. Panagis, E. Sakkopoulos, A. Tsakalidis. Personalization Techniques for Web Search Results Categorization, in the proceedings of the 2005 IEEE International Conference on e-Technology, e-Commerce and e-Service (IEEE EEE 2005), 29 March - 1 April 2005 in Hong Kong, China, pp. 148-151 P. Adamopoulou, D. Kanellopoulos, E. Sakkopoulos, A. Tsakalidis, Semantic learning interventions using web services technology, in the proceedings of IASTED 2005 International Conference Web Based Education, (WBE 2005), Web Based Teaching and Learning Technologies track, 2/2005, pp. 528-533 Ch. Makris, Y. Panagis, E. Sakkopoulos and A. Tsakalidis, An Algorithmic Framework for Adaptive Web Content (2005), in the International Workshop on Adaptive and Personalized Semantic Web in conjuction with the Sixteenth ACM Conference on HyperText and Hypermedia, proceedings published by Springer Verlag, Studies in Computational Intelligence, S. Sirmakessis ed., Vol. 14, 2006, pp. 1-10

535 529