OntoBlog: Linking Ontology and Blogs Aman Shakya Vilas Wuwongse Hideaki Takeda National Institute of Asian Institute of Technology National Institute of Informatics Klong Luang, Pathumthani Informatics 2-1-2 Hitotsubashi, 12120 2-1-2 Hitotsubashi, Chiyoda-ku Thailand Chiyoda-ku Tokyo 101-8430, Japan vw@cs.ait.ac.th Tokyo 101-8430, Japan shakya_aman@nii.ac.jp takeda@nii.ac.jp ABSTRACT navigating blog entries is difficult in traditional blogging. Semantic blogging attempts to enhance traditional blogging Semantic blogging attempts to address these issues by pro- by using Semantic Web technologies. Blog entries are se- viding a well defined structure to blog entries in the form of mantically enriched by metadata. However, authoring meta- metadata. It combines desirable features of both blogging data is not easy for normal users. Currently semantic blog- and the Semantic Web and extends blogging to informal ging only offers limited semantic capabilities. It is still diffi- knowledge management [5, 4]. Semantic blogging may also cult to navigate through semantically related entries, search be viewed as annotation to blog entries. We may annotate and organize relevant blog entries. OntoBlog attempts to blog entries with metadata based on some standard vocab- solve these issues by linking blogs to existing ontology main- ulary. However, manual entry of metadata is cumbersome tained using available ontology management environment. to the users and error-prone. Moreover, semantic blogging OntoBlog is a prototype semantic blogging system which should not only be about publishing metadata on the Web employs semi-automatic semantic annotation of blog entries but also linking related blog entries together based on se- using ontology instances. Blog entries are automatically mantics. mapped to related instances using language processing tech- niques. The rich structure of ontology with different se- In this paper we propose OntoBlog, a new semantic blog- mantic relations, enhanced by inference, can enable useful ging prototype which links blog entries to an existing on- semantic capabilities. Semantic navigation allows users to tology and instances. It allows semi-automatic annotation navigate through each blog entry to semantically related of blog entries with instances of the ontology. Any domain blog entries. Semantic search can be employed in blogs. ontology structured with various semantic relations may be Semantic aggregation collects blog entries relevant to the used and further enhanced by inference. Semantic annota- topic of interest and organizes them meaningfully. A pro- tion can help us to retrieve relevant resources; categorize totype for computer department domain ontology has been and organize contents and navigate meaningfully [11, 13, implemented. 20]. OntoBlog attempts to provide an integrated platform to facilitate publication, semantic annotation and informa- Categories and Subject Descriptors tion utilization. It demonstrates possibilities for semantic H.4.3 [Information Systems Applications]: Communi- capabilities in blogs by the use of an existing ontology. On cations applications—information browsers the other hand, blogging can be used to obtain feedback from the users for the improvement and maintenance of the ontology. General Terms Design In Section 2 we discuss about annotation and semantic an- notation in blogs. In Section 3 we propose OntoBlog1 as an Keywords integrated environment. A motivating scenario is described Blogging, ontology, Semantic Web, semantic annotation in Section 4. Section 5 describes implementation of the sys- tem. Some services offered by the system are discussed in 1. INTRODUCTION Section 6. In Section 7, we discuss some initial experiences Blogging has become very popular on the Web and is grow- testing the system. Section 8 describes some related works. ing rapidly [21]. Blogs make publishing information on the Finally, we conclude and point out future works in Section Web very easy for any user. But filtering, organizing and 9. 2. ANNOTATION Annotations2 are comments, notes, explanations, or remarks attached to any document or a selected part of the docu- ment. Annotation metadata can be used not only for de- scribing content, but also to organize and classify it [11]. 1 An online demo can be found at http://dutar.ex.nii.ac.jp/ontoblog/blog/default/ 2 http://www.w3.org/2001/Annotea/ Annotation can also help in information retrieval process [13, 20] and organizing search results [14]. 2.1 Automatic Semantic Annotation Annotation that references an ontology has been termed se- mantic annotation. Semantic annotation can enhance in- formation retrieval and improve interoperability [23]. With manual annotation, the user has the burden of creating an- notations. Providing useful annotation may also depend on the willingness of stakeholders [7]. Issues may also arise about the authenticity and quality of manual annotations. Automatic or semi-automatic annotation with pre-existing information can help in solving these issues. Automatic an- notation systems may provide suggestions, but still require Figure 1: Example scenario intervention by user; or acquire annotations automatically on a large scale. Uren et al. [23] describe different ways to provide automatic support for annotation: Integrated authoring. Semantic annotation is integrated with the authoring environment of the blog which helps the author to annotate entries easily at the time of blogging. • Wrappers can exploit the structure of web pages to identify pieces of information for annotation. Semi-automatic annotation. The system automatically • IE (Information Extraction) systems using supervised discovers related instances when blog entries are added and / unsupervised learning provides suggestions to the author. • Natural language processing Feedback for ontology maintenance. The users may suggest new concepts and instances if the system does not 2.2 Semantic Annotation in Blogs contain appropriate ones related to the blog entry. It can OntoBlog is an application of semantic annotation to blogs. be used as a feedback channel which would be helpful to the Blogs are somewhat different from other web information administrator to improve and maintain the knowledge base sources. Blog entries are self-contained snippets [4] of infor- up-to-date. mation or small contents [19]. We may consider a blog entry as a single unit of information. A characteristic of blogs is Integrated services. The system demonstrates how se- dynamic publishing. Blogs are user-oriented and provide an mantic capabilities may be incorporated to utilize blog con- easy mechanism for frequent publication. Therefore, there tents properly, by some example services. Semantic naviga- are some considerations for the application of semantic an- tion and search services are provided to retrieve and browse notation to blogs. related blog entries. Organization of related blog entries is provided by exploiting semantic relations in the ontology. Integrated authoring environment. Blogging provides an easy platform for publishing. Annotation would become easy if we can integrate it with the publishing platform. 4. EXAMPLE SCENARIO As an example domain, we consider the case of a com- An integrated environment providing a single point of entry puter department of a university. The department main- interface for publication and annotation has been pointed tains an ontology with concepts like course, topic, teacher, out as a requirement for semantic annotation systems by research, etc. The ontology is maintained by the ontology Uren et al. [23]. engineer or administrator using available ontology manage- ment software. The knowledge base has been populated Automation. Automation makes the process of annota- with instances of courses, topics, teachers, etc. The depart- tion fast and easy for the blogger. Utilizing existing ontol- ment also maintains a community blog as illustrated in Fig. ogy instances for annotation can be considered as automatic 1. Registered users can easily publish entries on the blog. metadata creation for blog entries. When publishing or updating a blog entry, the system auto- matically suggests instances related to the blog entry. The Integrated services. As blog entries are scattered in the user may accept the suggestions or modify some choices as form of small discrete entries, it becomes essential to provide appropriate. If a related instance or concept is not shown by some services to relate these pieces of information together the system, the user may type in appropriate instance name and present them to the user as an organized collection. as suggestion and select the proper concept, or suggest a new concept. The list of suggestions posted by users can 3. THE ONTOBLOG INTEGRATED PLAT- be viewed by the administrator. He/she can evaluate the FORM suggestions of various users and make appropriate additions We propose OntoBlog as an integrated authoring platform or improvements to the ontology. which links existing ontology to blogs. It incorporates the following features into blogging. The users can access the blog entries effectively with the help of semantic capabilities provided. When a blog entry is Semantic annotation. We propose to use existing ontolo- viewed, semantic navigation links are shown as related links. gies and instances to semantically annotate blog entries. Suppose our blog entry is related to databases. Then we will Figure 2: System architecture Figure 3: OntoBlog interface have a link for the ‘Database’ course and further links like - taught by ‘Prof. Vilas Wuwongse’ and has prerequisite ‘Data structures’. When the user clicks on ‘Data structures’, all is stored in the RDF metadata store as described in our blog entries related to that course are listed. Search results other work [22]. The RDF metadata store uses a MySQL are augmented by semantic search. Further, if the user is in- relational database to store RDF metadata. This provides a terested in some topic, he/she may use semantic aggregation scalable storage unlike using a single RDF file for all meta- to gather blog entries relevant to the topic and organize the data (as in [5, 4]). The Jena framework5 has been used to collection. Instances are shown inter-related to each other as manage operations on the RDF metadata store. Blog entries a directed graph. Implicit relations may also be revealed by and metadata can be entered by several users thus harness- enabling inference. The graph serves as a table of contents ing the collective intelligence in the community. Further, for the topic of interest. publishing metadata in RSS feeds makes way for aggrega- tion of information from multiple blogs. 5. IMPLEMENTATION 5.2 Ontology and Inference The system architecture is shown in Fig. 2. The system is The ontology contains various concepts from the domain built upon a blogging infrastructure backed up by an RDF and a wide variety of relations, not just a topic hierarchy. A metadata store. The Blojsom blogging platform has been simple example ontology of a computer science department used in the system. Blojsom3 offers extensibility with plug- was constructed for testing the system. The ontology was in architecture. Users post blog entries to the system. The created referring to the SHOE Computer Department On- blog entries may also contain some metadata. The system tology6 . However, only few parts of the ontology have been has a text search engine which indexes the blog entries and used including concepts and relations depicted in Fig. 4. performs keyword based search for user queries. Blog entries are linked to related ontology instances by the blog-ontology Instances of the ontology are populated in the knowledge linking component. The component automatically suggests base. We can also expect the availability of suitable knowl- related instances for blog entries and saves the selected an- edge bases. There are many knowledge bases in various do- notations in the metadata store. The inference engine can mains about real world entities being maintained by sev- deduce implicit relations between instances. Search results eral groups and enterprises [13, 20]. The knowledge base are augmented with related blog entries by finding linked maintained in a server can easily be shared by multiple dis- related instances from the ontology. All the related blog tributed blogs. entries obtained are finally organized meaningfully into a navigable collection by the organizer based on the structure Various mature ontology management software packages are of the ontology. The system also exports blog entries in RSS available which can be used to create and maintain the format with embedded metadata. knowledge base. Protégé7 has been used to create and main- tain the ontology and instances. 5.1 Publishing Metadata Besides normal text contents, semantic blog entries may also The OWL Micro reasoner from the reasoning subsystem contain metadata as shown in Fig. 3. In our test installation in Jena has been used for inference. It can be replaced we used some publication types from the SWRC ontology4 easily by other reasoners if more powerful inference is de- for the metadata schema. We can use any other metadata sired. The example ontology uses some axioms for inference. provided that an appropriate schema is available. Metadata 5 http://jena.sourceforge.net/inference/index.html 3 6 http://blojsom.sourceforge.net/ http://www.cs.umd.edu/projects/plus/SHOE/onts/cs1.1.html 4 7 http://ontoware.org/projects/swrc/ http://protege.stanford.edu/ Figure 4: A part of a computer department ontology For e.g. - “for course and has topic are inverse relations”, “is broader than and is narrower than are inverse relations”, “teaches and taught by are inverse relations”, “has prerequisite and is broader than are transitive”, etc. Figure 5: Example Blog-ontology linking 5.3 Blog-Ontology Linking The system links blogging to an existing ontology system remarks. The list of suggestions posted by various users by semi-automatic semantic annotation. Annotation can be can be accessed by the administrator on the blogging sys- automated by language processing of the blog entries. Lan- tem itself. The feedback thus collected is useful for the ad- guage processing may not be as sophisticated as techniques ministrator to maintain the ontology up-to-date by adding like information extraction and wrapper mechanisms. How- missing concepts and instances in the ontology or refining ever, simple lexical analysis can be very fast and quite ef- them. The administrator uses an existing ontology manage- fective for annotating by existing instances [7]. The Porter ment software to make updates based on the suggestions. stemming algorithm has been used (as in [5, 4]). Stem- The suggestion also shows the permalink of the concerned ming is a widely used technique in information retrieval and blog entry so that it can easily be annotated with the newly though it is a simple method, produces quite good results. defined instance. The system provides automatic suggestions to the user for annotation. The user can easily modify the options if some 6. SEMANTIC SERVICES suggestions are not appropriate. Automatic annotation can- Semantic annotation of blog entries allows us to relate dif- not be perfect even with other available sophisticated tech- ferent blog entries using the structure of the ontology as niques. Moreover, relevance is a subjective matter and not illustrated in Fig. 6. In the figure, instances (I1-I7) in the possible for perfect automatic judgment. Providing sugges- ontology are represented by different shapes, each shape rep- tions to the user keeps the system flexible instead of making resenting a concept. Instances are connected to each other it totally automatic and rigid. by different relations (indicated by the solid arrows). Link- ing blog entries to ontology serves to link related blog entries A “keywords” element has been added to each concept in the implicitly. Blog entries (A to F) are annotated by the on- ontology. For each instance, the “keywords” element con- tology by linking them to the instances, as shown by the tains a collection of words related to that instance. When- dash-dotted lines. Blog entries ‘A’ and ‘B’ are related to ever an entry is added or updated, the “keywords” for each each other because they are both mapped to the same in- instance are stemmed and matched against the stemmed stance ‘I1’. A blog entry may be related to multiple in- blog entry. If any of the “keywords” is found in the stemmed stances (like ‘E’ related to ‘I6’ and ‘I7’). Instance ‘I1’ is text, the blog entry is considered as related to that instance. related to ‘I2’. Hence, blog entry ‘A’ is indirectly related to Instances related to the blog entry are automatically discov- ‘C’, which has been mapped to ‘I2’. Instances may also be ered and suggested to the user. The discovered relations, linked by implicit relations (shown by dashed arrow) that after user’s approval, are encoded in the related-to property, can be discovered by inference. Instance ‘I4’ is related to as shown in Fig. 5. In the example, the blog entry with ‘I6’ by an inferred link. Thus, blog entry ‘D’ (mapped to permalink “RMI.html” has been linked to ‘I4’) is related to ‘E’ (mapped to ‘I6’). “Java Technologies” and“Remote Method Invocation”in the ontology. Section 6 further illustrates how blog entries are linked to ontology instances to enable semantic capabilities. 6.1 Semantic Navigation Semantic navigation helps the user to browse through re- lated blog entries. The following algorithm is used for the 5.4 Feedback for Ontology Maintenance purpose. In case some related instances or concepts are not defined in the ontology, users may suggest a suitable instance or concept. The system provides a web-based form for new 1. While viewing a blog entry B, get all related instances suggestions along with the automatic suggestions for blog- from the Blog-Ontology relations. Let I be the set of ontology linking described in the previous section. The user these instances. may enter the name for the new instance, select the appro- priate concept (or suggest a new concept) and post some 2. For each instance i in I search attempts to augment and improve traditional search results using data from the Semantic Web. Guha et al. [9] have presented extensive research on semantic search along with sophisticated implementation. We do not intend to re- produce or extend the research on semantic search. Rather OntoBlog just provides a simple implementation of semantic search for demonstration of its applicability. The algorithm is outlined below. 1. Find all instances in the ontology with keywords match- ing the search text. Let the set of instances obtained be I. 2. Set final list of instances L = I. Figure 6: Example Blog-ontology linking 3. For each instance i in I (a) Let J be the set of instances directly related to i. (b) L = L union J 4. For each instance in L, get all related blog entries from Blog-Ontology relations. 5. Return the total blog entries as result. Semantic search may be particularly useful when text search alone does not produce enough results. It may be better to have more results than just few accurate results. Seman- tic search can be enabled or disabled using configuration settings. Further, the depth of semantic search can also be controlled. If semantic search produces excess results in some cases, the search depth can be decreased or semantic search can be disabled altogether. Figure 7: Semantic navigation 6.3 Semantic Aggregation Semantic aggregation can be introduced in the system to col- (a) Find all relations R involving i lect and organize search results relevant to a topic of inter- est. A simple algorithm for semantic aggregation is outlined (b) For each relation r in R, find instances related to below. i by r. 3. Render all instances with hyperlinks. 1. Get all relevant blog entries from search. For example, suppose we view a blog entry B about“Database 2. Find the set of instances S from the Ontology linked Programming”. The blog entry may be connected to, I = to each blog entry. “computer programming”, “databases”, “software develop- ment”, “Prof. Takeda” .... “Computer programming” may be 3. Find all relations between the instances in S involved in R = “is taught by”, “has prerequisite”,.... Thus, 4. Visualize the related instances as directed graphs. there may be links like [computer programming] For example, let the search results for some search contain – is taught by – [Prof. Takeda] following blog entries. – has prerequisite – [databases],etc. entry A - related to - Computer programming, data struc- Clicking on [databases] will lead to the blog entries related ture to databases. When a blog entry is opened, the semantic entry B - related to - XML, database, Prof. Takeda navigation links are shown in a collapsible “Related to” block entry C - related to - Java, OOPS, etc (shown in Fig. 7). Then, S = Computer programming, data structure, XML 6.2 Search .... The instances in S may be related with each other by The system provides indexed text search and metadata search. relations like prerequisite, broader/narrower, taught by, is These are further augmented by semantic search. Semantic related to, etc. Table 1: Semantic search results Figure 8: Semantic aggregation Prefuse8 has been used to visualize the graphs. GraphML9 has been used to represent the semantic aggregation graphs. Semantic aggregation is depicted in Fig. 8. The user runs se- mantic aggregation by searching on a topic of interest. The search results are listed on the right-hand side frame. Re- lated instances from the ontology are aggregated and visual- ized on the left-hand side frame. The relation type between instances is identified by the color of the link and shown in an index. When a node is clicked, blog entries related to results from simple language processing like stemming alone. that node are displayed on the right-hand side frame. “Keywords” defined to identify an instance, also affects the process. It is difficult to exhaustively list keywords related 6.4 Other Features to any instance. The system generates RSS feeds with embedded metadata. As demonstrated in our other work [22], RSS feeds from Semantic search. Semantic search was tested with a num- multiple blogs in the community can be aggregated. How- ber of search texts related to some course or topic as shown ever, the new ontology-based features are yet to be applied in Table 1. The table verifies that semantic search can pro- on such aggregate from multiple blogs. The system also duce additional relevant results not produced by text search provides a bi-directional commenting mechanism, employs alone hence increasing the recall of the system. bookmarklet for easy publishing and generates FOAF pro- file for interconnecting the community as described in [22]. Semantic aggregation. Semantic aggregation was tested only to see if relevant instances are retrieved during the pro- 7. SOME EXPERIENCES cess. The precision was roughly found to be about 77.83%. But semantic aggregation provides much more than simply We attempted to test some features of the system by con- ducting simple experiments. The data for populating in- retrieving relevant instances. stances in the ontology was mostly adapted from the web- User feedback. Many features of the system are subjec- site of the Computer Science and Information Management tive in nature and difficult to evaluate by experiments. Tra- department of the Asian Institute of Technology10 . Data ditional information retrieval metrics like recall and preci- about 15 courses, 93 topics, 8 faculty members, 10 researches and 15 universities have been entered. About 100 dummy sion are not suitable enough for evaluation of a system like blog entries related to different courses, topics, researches, this [13, 20]. So user feedback about the system was also collected. The questionnaires had questions about the effec- etc were populated manually. 10 students from the depart- tiveness of semantic navigation, semantic search, inference, ment used the system to help in experiments. Statistical semantic aggregation, metadata search and RSS aggrega- treatment was not possible at this stage due to the limited tion. The users had good response for most features. How- number of users. ever, it was not so easy for the users to judge relevance of Automatic blog-ontology linking. While populating semantic search results because the results were not directly related to the query. blog entries, right and wrong suggestions were noted. The relevance of the suggestions to the blog entries was judged by human subjects. The result showed about 84% right sugges- 8. RELATED WORK tions and 16% wrong suggestions. We cannot expect perfect A number of works have been done in semantic blogging. 8 The Semantic Blogging Demonstrator11 uses a category tree http://prefuse.sourceforge.net/ 9 based on ‘broader than/narrower than’ relations [5, 4] to cat- http://graphml.graphdrawing.org/ 10 11 http://www.cs.ait.ac.th/ (retrieved April, 2006) http://www.semanticblogging.org/semblog/blog/default/ egorize blog entries. OntoBlog uses an ontology with wider and linguistic rules to identify instances. Magpie depends on variety of concepts and relations, rather than a simple taxon- external service providers for providing semantic services. omy, and enables powerful inferences. It also allows utilizing KIM [13, 20] uses an ontology with a pre-populated knowl- rich knowledge base of instances maintained using existing edge base of instances. IE techniques have been employed ontology management environments. It uses OWL which for the recognition of named entities in documents. It also is a more powerful ontology language than SKOS used by introduces indexing and retrieval based on named entities. the demonstrator. OWL allows us to define several types of relations and offers powerful inferences which may be needed in future. The demonstrator mainly emphasizes se- 9. CONCLUSIONS AND FUTURE WORK mantic view, navigation and query. However, Tree brows- In this paper we proposed OntoBlog, a prototype semantic ing provided for semantic navigation is already a feature blogging system which semi-automatically annotates blog of many blogs [3]. Facet browsing is more like metadata entries with instances of an ontology. It links together the search. OntoBlog offers a different way of semantic navi- well established technologies of blogging and ontology man- gation by providing related links through each blog entry agement. The system is helpful in organizing the contents in which is more intuitive way of browsing. Search has been blogs through semantic capabilities and also receiving feed- enhanced to explore semantic relations. Furthermore, On- back from the users for ontology maintenance to some ex- toBlog organizes the search results by semantic aggregation. tent. The deep semantic structure of ontology provides a Works like [12], [19] and [16, 18, 17] are generic. We used rich way of classifying and organizing relevant entries and computer science department domain (also used by Man- can enable better navigation and search capabilities in blogs. grove [15]) just as a simple example. OntoBlog can be used OntoBlog provides an integrated platform for publishing and in any domain provided that a knowledge base is maintained. information utilization in blogs. The demonstrator provides a category chooser functionality which works based on simple language processing. We also We can explore mechanisms for the decentralized creation used a similar technique for automatic annotation by ontol- of the ontology in future. Collaborative techniques like folk- ogy instances. sonomy or semantic wiki could be utilized. The system is currently a community blog for the concerned domain. Karger and Quan [12] extended Haystack12 to enable users Features like semantic navigation and search are yet to be to view cross-blog reply graphs and track conversation in employed across multiple blogs. The system can be made multiple blogs. Semblog [19] annotates content using FOAF more powerful by introducing different types of inference metadata of users and exports using extended RSS. Our and inference engines. Mature semantic search systems can other work [22] facilitates sharing of bibliographic informa- be incorporated. Information retrieval mechanisms can be tion in a social network based on extended RSS. Möller et used to produce ranked search results. The system may be al. [16] identify structural and content-related metadata in extended to handle complex hierarchical metadata schema. blogging. The SIOC ontology [2] has been used for struc- Language processing with stemming used for the demonstra- tural metadata. FOAF, vCard, BibTex/SWRC, iCalendar, tion prototype is quite basic. It can be made more accurate etc. have been used for content metadata. In our case, by handling lexical, semantical and syntactical variations [1]. existing commenting mechanism generates some structural Lexical variations can be dealt with technologies like Word- metadata. Though SWRC has been used as an example, any Net. Sophisticated automatic annotation mechanisms using content metadata can be used. In addition, the metadata IE techniques can be incorporated to make the automation produced by automatic semantic annotation can be consid- more robust. Supervised or unsupervised learning can fur- ered as categorization metadata. semiBlog [16, 18, 17] em- ther augment such IE techniques. It would also be good to phasizes generating metadata by utilizing data on the user’s support semantic blogging clients like semiBlog [16, 18, 17] desktop. But the user still has to search metadata assuming to utilize data from the user’s desktop. that it exists in his/her desktop. More abundant metadata, of better quality, may be available in existing knowledge bases than one’s desktop. 10. ADDITIONAL AUTHORS Additional authors: Ikki Ohmukai (National Institute of In- formatics, email: i2k@nii.ac.jp) Uren et al. [23] present a detailed survey of annotation frame- works and semantic annotation tools and analyze them on the basis of a number of requirements. A large body of re- 11. REFERENCES search on semi-automatic semantic annotation already exists [1] A. Arampatzis, T. van der Weide, C. Koster, and including significant works like S-CREAM [10] and extrac- P. van Bommel. Linguistically motivated information tion ontologies [6]. Our attempt is to demonstrate the appli- retrieval. In A. Kent, editor, Encyclopedia of Library cation of semantic annotation in blogs, not to build a sophis- and Information Science, volume 69. Marcel Dekker, ticated annotation system satisfying all the requirements Inc., New York, Basel, December 2000. mentioned in [23]. OntoBlog satisfies some requirements like [2] J. G. Breslin, A. Harth, U. Bojars, and S. Decker. automation, integrated environment, document-annotation Towards semantically-interlinked online communities. consistency and separate annotation storage. Magpie [7, 8] In The Semantic Web: Research and Applications, automatically creates a semantic layer over web documents volume 3532 of Lecture Notes in Computer Science, and links instances identified in the document to relevant on- pages 500–514. Springer Berlin / Heidelberg, 2005. tological instance/class. It uses simple lexicon-based parsing [3] S. Cayzer. Swad-europe deliverable 12.1.4: Semantic blogging - lessons learnt. 12 http://haystack.lcs.mit.edu http://www.w3.org/2001/sw/Europe/reports/demo 1 report/. [4] S. Cayzer. Semantic blogging and decentralized [15] L. McDowell, O. Etzioni, S. D. Gribble, A. Halevy, knowledge management. Communications of the H. Levy, W. Pentney, D. Verma, and S. Vlasseva. ACM, 47(12):48–52, December 2004. Mangrove: Enticing ordinary people onto the semantic [5] S. Cayzer. Semantic blogging: Spreading the semantic web via instant gratification. In The Semantic Web - web meme. In Proceedings of XML Europe 2004, pages ISWC 2003, volume 2870, pages 754–770, Florida, 18–21, Amsterdam, Netherlands, April 2004. USA, 2003. Springer Berlin / Heidelberg. [6] Y. Ding, D. W. Embley, and S. W. Liddle. Automatic [16] K. Möller, U. Bojars, and J. G. Breslin. Using creation and simplified querying of semantic Web semantics to enhance the blogging experience. In The content: An approach based on information-extraction Semantic Web: Research and Applications, volume ontologies. In Proceedings of the First Asian Semantic 4011 of Lecture Notes in Computer Science, pages Web Conference (ASWC 2006), volume 4185 of 679–696. Springer Berlin / Heidelberg, 2006. Lecture Notes in Computer Science, pages 400–414, [17] K. Möller, J. G. Breslin, and S. Decker. semiblog - Beijing, China, 2006. Springer. semantic publishing of desktop data. In 14th [7] M. Dzbor, J. Domingue, and E. Motta. Magpie - Conference on Information Systems Development towards a semantic web browser. In The Semantic (ISD2005), Karlstad, Sweden, August 2005. Web - ISWC 2003, volume 2870 of Lecture Notes in [18] K. Möller and S. Decker. Harvesting desktop data for Computer Science, pages 690–705, Florida, USA, 2003. semantic blogging. In Proceedings of the 1st Workshop Springer-Verlag Berlin Heidelberg. on The Semantic Desktop - Next Generation Personal [8] M. Dzbor, E. Motta, and J. Domingue. Opening up Information Management and Collaboration magpie via semantic services. In The Semantic Web - Infrastructure at ISWC2005, Galway, Ireland, ISWC 2004, volume 3298 of Lecture Notes in November 2005. Computer Science, pages 635–649. Springer Berlin / [19] I. Ohmukai and H. Takeda. Semblog: Personal Heidelberg, 2004. knowledge publishing suite. In Proceedings of WWW [9] R. Guha, R. McCool, and E. Miller. Semantic search. 2004 Workshop on the Weblogging Ecosystem: In Proceedings of the twelfth international conference Aggregation, Analysis and Dynamics, New York, USA, on World Wide Web, pages 700–709, Budapest, 2004. Hungary, 2003. ACM Press New York, USA. [20] B. Popov, A. Kiryakov, A. Kirilov, D. Manov, [10] S. Handschuh, S. Staab, and F. Ciravegna. S-cream - D. Ognyanoff, and M. Goranov. KIM-Semantic semi-automatic creation of metadata. In 13th Annotation Platform. In 2nd International Semantic International Conference on Knowledge Engineering Web Conference (ISWC2003), volume 2870 of Lecture and Knowledge Management (EKAW02), Madrid, Notes in Computer Science, pages 834–849. Springer, Spain, October 2002. 2003. [11] J. Kahan, M. R. Koivunen, E. Prud’Hommeaux, and [21] L. Rainie. The state of blogging. R. R. Swick. Annotea: an open rdf infrastructure for http://www.pewinternet.org/PPF/r/144/report display.asp, shared web annotations. In Proceedings of the 10th 2005. International World Wide Web Conference, pages [22] A. Shakya, H. Takeda, I. Ohmukai, and V. Wuwongse. 623–632, Hong Kong, May 2001. A publication aggregation system using semantic [12] D. R. Karger and D. Quan. What would it mean to blogging. In G. Li, Y. Liang, and M. Ronchetti, blog on the semantic web? Journal of Web Semantics, editors, The Semantic Web - ASWC 2006 Workshops 3(2):147–157, 2005. Proceedings, pages 55–62, Beijing, China, September [13] A. Kiryakov, B. Popov, D. Ognyanoff, D. Manov, 2006. Jilin University Press. A. Kirilov, and M. Goranov. Semantic annotation, [23] V. Uren, P. Cimiano, J. Iria, S. Handschuh, indexing, and retrieval. In 2nd International Semantic M. Vargas-Vera, E. Motta, and F. Ciravegna. Web Conference (ISWC 2003), volume 2870 of Lecture Semantic annotation for knowledge management: Notes in Computer Science, pages 484–499, Florida, Requirements and a survey of the state of the art. USA, 2003. Springer-Verlag Berlin Heidelberg. Journal of Web Semantics: Science, Services and [14] M.-R. Koivunen. Annotea and semantic web Agents on the World Wide Web, 4(1):14–28, 2006. supported collaboration. In ESWC 2005, UserSWeb workshop, 2005.