I posed my independent study report for others and I think u may get help on this. My independent study based on the semantic search.
Semantic Search
Wijekoon H.M.I.G.A.B. (064107U)
Faculty of Information Technology
University of Moratuwa
Abstract: In this review paper I try to present about ‘Semantic Search’ which is built on Semantic Web technologies. I provide an overview of RDF, TAP which is the application framework upon which the Semantic Search is built and five distinct research directions in Semantic Search presenting today. Those paths are augmenting traditional keyword search with semantic techniques, basic concept location, complex constraint queries, problem solving and connecting path discovery. I also discuss some general issues related to searching and the Semantic Web and Semantic Search engines currently present.
1) Introduction
In web 3.0 come up with the concept of Semantic Web and Semantic Search [1]. Semantic Search will be the major searching mechanism and application run on the Semantic Web. Activities such as Web Services and the Semantic Web are working to create a web of distributed machine understandable data. Based on that web searching will be more familiar to the web consumers than today. Today we are using “phases” to find out the web resources. Like google, yahoo, msn, etc do a great job to find the resources and still there are some unsatisfactory things are presenting. One of the thing is the resources which are finding by the search engines are depend on the phrase what we are queering. So still there are draw backs and to overcome those things people think a searching mechanism and the answer for that is Semantic Search.
Following section you will give a brief description about how my In “Overview” section I mention about broader definitions regarding to Semantic web and Semantic Search. In “Semantic Searching” section I present broader reviewing about the topic. In hear I present the history of the Semantic Search, basic foundation of the Semantic Web that is RDF and what are the research paths of Semantic Searching and their behind concept. Hear I also consider about the relation between AI techniques with Semantic Search. In next section “Applications of Semantic Searching” I tell about the industrial Semantic Search engines which are now already built and currently ongoing developing ones. Then I mention about future directions in Semantic Searching in “Future Direction of Semantic Searching” section. Next in “Discussion” section I give a summary of all what I found and where I will be moving next. Then is the most critical section of the review paper which is my contribution for the topic of the review paper. This section is naming as “Contribution” and then “Acknowledgement” section. In hear I thank to people who are supported for my independent study. Finally “Reference” sections for state the references.
3) Overview
Semantic Search is working on the Semantic Web. So first it’s better if we have an idea about the Semantic Web. Semantic Web is the extension of the World Wide Web that enables people to share content beyond the boundaries of applications and websites [2]. It has been described in rather different ways. One of the definitions is saying it as a web of data or merely as a shifting of our natural paradigm in our daily use of the Web [3]. The word semantic stands for the meaning of. The semantic of something is the meaning of something. The Semantic Web is a web that is able to describe things in a way that computers can understand.
- The Kandy is one of a popular city in Sri Lanka.
- “Dalada Maligawa” is the Kandy town.
- The tooth relic reside in hear.
Sentences like these can be understood by people. But how can they be understood by computers? Statements are built with syntax rules. The syntax of a language defines the rules for building the language statements. But how can syntax become semantic? This is what the Semantic Web is all about. Describing things in a way that computers applications can understand. The Semantic Web is not about links between web pages. The Web was designed as an information space, with the goal that it should be useful not only for human-human communication, but also that machines would be able to participate and help. One of the major obstacles to this has been the fact that most information on the Web is designed for human consumption. The Semantic Web approach develops languages for expressing information in a machine processable form.
As with the WWW, the growth of the Semantic Web will be driven by applications that use it. Semantic Search is an application of the Semantic Web to search. Search is both one of the most popular applications on the Web and an application with significant room for improvement. This is process what we are doing in normal searching. User queries submitted to web search engines are not always informative enough for retrieving the related pages to the user intention. The main problem is that users may not know the best query items they should enter to get the most related web pages to their intentions. They may not be familiar with the specific keywords in that domain knowledge. A user may remember only a part of the phrase that he/she wants to use in the query string. Sometimes the user does not know how to order the keywords (most web search engines are sensitive to the order of the keywords) or even does not know the correct spelling of a specific keyword in the query string. A novice user sometimes sends an imperfect query and scans the returned web pages (even reads a number of the returned documents) to prepare a more precise query by finding new related keywords in the documents. To overcome these types of problems the answer is Semantic Search. In Semantic Search it can basically divide it five types of searching. They are augmenting traditional keyword search with semantic techniques, basic concept location, complex constraint queries, problem solving and connecting path discovery.
So whole in this review paper I discuss about the Semantic Search and I also discuss some little bit about Semantic Web be course we can not speak about Semantic Search without Semantic Web.
04) Semantic Searching
When we are considering the Semantic Search first we are looking at the Semantic Web. So when looking at a possible developing way of a universal Web (Semantic Web) we have to think about the principle of a common model of great generality [5]. Only when the common model is general can any prospective application be mapped onto the model. The general model is the Resource Description Framework.
The Resource Description Framework (RDF) is an extremely flexible technology, capable of addressing a wide variety of problems. Because of its enormous breadth, people often come to RDF thinking that it’s one thing and find later that it’s much more.
According to the W3C (World Wide Web Consortium) definition of the Resource Description Framework (RDF) is a language designed to support the Semantic Web, in much the same way that HTML is the language that helped initiate the original Web. RDF is a framework for supporting resource description, or metadata (data about data), for the Web. RDF provides common structures that can be used for interoperable XML data exchange.
When we are considering the RDF Concepts, Abstract Syntax and the RDF Semantics documents provide the fundamental framework behind RDF. The underlying assumptions and structures those make RDF unique from other metadata models (such as the relational data model). These documents provide both validity and consistency to RDF. So it provides a way of verifying that data structured in a certain way will always be compatible with other data using the same structures [6]. The RDF model exists independently of any representation of RDF, including RDF/XML. RDF’s purpose is fairly straightforward: it provides a means of recording data in a machine-understandable format, allowing for more efficient and sophisticated data interchange, searching, cataloging, navigation, classification, and so on.
Then we need an infrastructure for applications on the Semantic Web. TAP is a system for easily publishing, discovering and consuming structured data [10]. Consequently, TAP’s architecture is defined by the data model (and format) for this data and the protocol for discovering and querying the data. The TAP software system consists of a number of loosely coupled modules, with the only commonality being the adherance to the data model, format and protocol. TAP provides a set of simple mechanisms for sites to publish data onto the Semantic Web and for applications to consume this data via a minimalist query interface called GetData. Already we described RDF, there are number of query languages develop for use as queering in RDF. Some of they are jena, Squish, DAML DQL and more generally for semi-structured data Lore, XQuery [9].
These query languages all provide very expressive mechanisms that are are aimed at making it easy to express complex queries. Unfortunately, with such expressive query languages, it is easy to construct queries that require a lot of computational resources to process. Consequently, just as no major Website provides a SQL interface to its back end relational database, we don’t expect sites, especially large ones, to use these query languages as the external interface to their data. What we need is a much lighter weight interface, one that is both easier to support and more importantly, exhibits predictable behavior. Predictable behavior is important not just for the service provider, but also the service client. A simple lightweight query system would be complementary to more complete query languages mentioned above. The lighter weight query language could be used for querying on the open, uncontrolled Web in contexts where a site might not have much control over who is issuing queries, whereas the latter is targeted at the comparatively better behaved and more predictable area behind the firewall. The lightweight query also does not preclude particular sites from aggregating data from multiple sites and providing richer query interfaces into these aggregations.
GetData is intended to be a simple query interface to network accessible data presented as directed labeled graphs. GetData is not intended to be a complete or expressive query language a la SQL, XQuery, RQL or DQL. It is intended to be very easy to build, support and use, both from the perspective of data providers and data consumers. We want to enable machines to query remote servers for data. Since SOAP provides a mechanism for performing RPC that is beginning to be widely accepted, GetData is built on top of SOAP [8].
By using the RDF technology and TAP infrastructure those have added improvements to the Semantic Web. So those enable the searching process easy. Semantic Search attempts to augment and improve traditional search results (based on Information Retrieval technology) by using data from the Semantic Web.
Traditional Information Retrieval (IR) technology is based almost purely on the occurrence of words in documents. Search engines like Google , augment this in the context of the Web with information about the hyperlink structure of the Web. The availability of large amounts of structured, machine understandable information about a wide range of objects on the Semantic Web offers some opportunities for improving on traditional search. Before getting into the details of how the Semantic Web can contribute to search, we need to distinguish between two very different kinds of searches. Already I explain about those two searches and they are
- Navigational Searches: In this class of searches, the user provides the search engine a phrase or combination of words which user expects to find in the documents. There is no straightforward, reasonable interpretation of these words as denoting a concept. In such cases, the user is using the search engine as a navigation tool to navigate to a particular intended document. We are not interested in this class of searches.
- Research Searches: In many other cases, the user provides the search engine with a phrase which is intended to denote an object about which the user is trying to gather/research information. There is no particular document which the user knows about that she or he is trying to get to. Rather, the user is trying to locate a number of documents which together will give him/her the information s/he is trying to find. This is the class of searches we are interested in [4].
Since the Semantic Web does not yet contain much information, in addition to the Semantic Search application, we have had to build the requisite portions of the Semantic Web to provide data for the Semantic Search applications. Let’s consider some features of Semantic Web environment.
The information about music, cars, tickets (and everything else) were stored in RDF files, intelligent web applications could collect information from many different sources, combine information, and present it to users in a meaningful way [7].
Information like this:
· Car prices from different resellers
· Information about medicines
· Plane schedules
· Spare parts for the industry
· Information about books (price, pages, editor, year)
· Who is who
· Dates of events
· Computer updates
There are so many other features too. But now it’s enough consider about Semantic Web and now we move to the our main topic and if need more feature description of Semantic Web those are describe with in those parts.
After I studying many review papers I find there are five main directions in Semantic Searching and those have own different behaviors compare with others. So let’s consider each one by one.
Augmenting Traditional Keyword Search with Semantic Techniques
Early research on semantic web enabled search deals with augmenting traditional text search with semantic techniques. This research direction differs significantly from the others presented later. Which is in the sense that it does not usually believe the bulk of the actual knowledge being sought to be formally annotated. Instead, ontological techniques are used in a multitude of ways to augment keyword search, whether to increase recall or precision. In the following, a variety of approaches is presented:
Many query expansion implementations utilized in keyword search make use of thesaurus ontology navigation as a step in query expansion. Particularly utilized is the largeWordNet ontology, defining synonym and meronym sets for words. These systems all function along the same basic scheme: first, the keywords are located in the ontology, then, various other concepts are located through graph traversal, after which the terms related to those concepts are utilized to either broaden or constrain the search. In [11] and [12], terms are expanded to their synonym and meronym sets using the boolean OR operations available in most search engines. In Clever Search[13], a particular meaning of a word in the WordNet ontology can be selected, resulting in the clarification text of that meaning being added to the search keywords via the boolean AND operator. In the ontology navigation phase, the implementations differ mostly in which properties of the ontology are navigated and which terms are picked.
A very simple manner of augmenting traditional keyword search results is taken in the “Semantic Search” interface [4] of the TAP infrastructure. Here, in addition to a traditional keyword search of TAP, targeted at a document database, the keywords are matched against concept labels in an RDF repository. Matching concepts are then returned in addition to the located documents. The paper also posits a continuation of the search similar to the one described in [13], where, if multiple concepts match the keyword, the user can select his intented meaning to constrain the search. Here, however, the idea is not to expand search terms, but to use some procedure to classify the actual documents as pertaining or not pertaining to concepts, and constraining results based on that semantic annotation. Some of research papers [15] describe algorithms for locating additional information relevant to a query given a starting set obtained via text search. First, traditional text search is applied into a document collection. Then, a process of RDF graph traversal is begun from the annotations of those documents. The intention is to find related concepts such as the writer of the document, the project the document refers to, etc. in a general manner.
The traversal is done by a spread activation algorithm, for the use of which the arcs in the ontology are weighed according to general interestingness. This is calculated by combining a specificity measure favoring unique connections in the knowledge base, and a cluster measure, which favors links between similar concepts. The CIRI[16] search system provides an ontological front-end to text search. The search is done through an ontology browser that visualizes the ontologies created for search as subsumption trees, from which concepts can be selected to constrain the search. The actual search is done through keywords annotated to these concepts, as well as subconcepts, utilizing a traditional text search engine and boolean logic. The actual search algorithm is in many ways similar to the query expansion algorithms discussed before. The main difference is in the user interface being based on direct ontological browsing, leaving out the first step of mapping a search keyword to the ontology.
Basic Concept Location
While much of semantic search research is directed at adding semantic annotations to data in order to improve search precision and recall on that data, there are other reasons for writing down information with formal semantics. Therefore, some research begins with the assumption of concepts, instances and relationships, and deals with the task of efficiently locating instances of these core semantic web datatypes. Usually, data on the semantic web is divided into two classes: ontological and instance data. The actual data the user is interested in are individuals belonging to a class, but the domain knowledge and relationships is described primarilty as class relationships in the ontology. This organization of data points to a natural way of locating information, exemplified for example in the SHOE search system[17]. In SHOE, the user is first given a visualization of the subsumption tree of classes in the ontology, from which he can choose the class of instances he is looking for. Then, the possible relationships or properties associated with the class are sought, and a form is presented that allows the user to constrain the set of instances by applying keyword filters to the various instance properties. When the properties point to objects, the target of the filtering will be the label of the referenced resource. Queries that can be formulated via this paradigm are e.g. “find all publications with a particular author name, from a particular project”. A similar approach is also taken in some versions of the SEAL portal tool[18]. The class subsumption-tree -based approach here is similar to the single facet search used in many Internet directories such as dmoz.org and Yahoo!. A more powerful paradigm is that of multi-facet search[19]. This is the approach behind the main search of the OntoViews[20]-based portals, and the SWED[21] directory portal. In multi facet search, multiple distinct views are provided into the data. For example, in the OntoViews -based MuseumFinland portal[22], where the information items are museum objects, the user is given views such as Object Material, Place of Manufacture and Context of Use. These views are created via ontology projection, utilizing also the various other hierarchical relationship trees and leaf relations usually inherent in ontologies besides class subsumption and membership. Here, the idea is that the user can start constraining their search from the view that is most natural to them. Additionally, constraints on the different views can be combined to create more complex queries, so the user can for example search for museum objects manufactured in China and used in Fishing. Additional implementations of the idea include the Longwell browser of the Simile project1, which differs in that it is restricted to flat views. In some versions of the OntoViews semantic portal creation tool [20], a concept called semantic auto completion [23] is utilized, which makes use of keyword search as a prelude to ontological navigation. The idea is taken to its furthest in the Veturi yellow pages service discovery portal [24], where the main interface of the portal opens with a keyword field. The keywords, however, are not linked directly to information items, but to ontological classes in the different views, from which semantic disambiguations can be made. The search then proceeds as a multi-facet search query.
Once the search has proceeded to the point where at least a single interesting instance is located, additional information can be retrieved via browsing. The process is analogous to the browsing of web pages linked together via hyperlinks. However, here the items shown are resources and the links between them are defined by their relations. In the simplest case, one concept is shown at a time, along with its properties derived straight from the RDF triples. If a property points to another resource and not a literal, then clicking on that property will browse to the referenced concept.
The authors of the Haystack information management tool[25, 26] base their user interface paradigm almost completely on browsing from resource to resource. This is argued by search behavior research[27], in which the authors posit that actually most searching is done via a process they call orienteering. Here, the premise is that searchers usually don’t actually themselves know or remember the specific qualities of what they are looking for, but have some idea of other things related to the sought item. The process of search is then a browsing experience in which the searcher looks for information resources that he knows are somehow related to the target, and from there locates additional information on the target resource until it can be located. An example in the article is of a person searching for a particular piece of documentation. Not remembering where it is stored, she only remembers that it was referenced to in some email message from a co-worker. She then scans through her mails in her inbox, remembering the co-worker who the mail was probably from, locates the correct message and from there extracts the location of the documentation. To facilitate locating points of entry for orienteering, Haystack provides a simple text search interface, based on the rationale that the things people remember about resources are probably their labels or phrases contained in them. The OntoViews-based portals also offer a browsing functionality between individual information items. This is realized through formalizing interesting RDF path patterns as Prolog rules, and linking to items at the endpoints of pattern-fitting paths beginning at the current item. This allows for linking complexly related items to each other, such as linking a person to all his or her distant relatives.
Complex Constraint Queries
Many kinds of complex queries can be formulated as locating a group of objects of certain types connected by certain relationships. In the semantic web, this translates to graph patterns with constrained object node and property arc types. An example would
be “Locate all toys manufactured in Europe in the 19th century, used by someone born in the 20th century”, where “toys”, “Europe”, “the 18th century”, “someone” and “the 19th century” are ontological class restrictions on nodes and “manufactured in”, “used by” and “time of birth” are the required connecting arcs in the pattern.
While such patterns are easy to formalize and query in the context of the semantic web, they remain problematic because they are not easy for users to formulate directly. Therefore, much of the research into complex queries has been on the level of user interfaces for creating such query patterns as intuitively as possible. [28] Presents GRQL, a graphical user interface for building graph pattern queries that is based on navigating the ontology. First, a class in the ontology is selected as a starting point. All properties defined as applicable to the class in the ontology are then given for expansion. Clicking on a property expands the graph pattern to contain that property, and moves selection to the range class defined for that property, e.g. clicking the creates property in an Artist class creates the pattern “Artist!creates!Artifact”, and moves focus to the Artifact class, showing the properties for that class for further path expansion. In addition to lenghtening the path, other operations can be performed on the query pattern. The pattern can be tightened to concern only some subclasses of a class, as in tightening the previous example to “Artist ! creates ! Painting or Sculpture”. In a similar way, property restriction definitions can be tightened into subproperties. More complex queries can be formulated by visiting a node created earlier and branching the expression there, creating patterns such as the one visually depicted
in figure 1 which could be used to find all artists that have either created paintings good enough to be exhibited at a museum, or any sculptures, as well as those sculptures, paintings and museums.
Another graphical query generation interface is described in [29]. Here, the user is given some pre-prepared domain-specific patterns to choose from as a starting point, which he can then extend and customize. The refinements to the query can be either additional property constraints to the classes, e.g. “Industry with sector Agriculture” or a replacement of a class in the pattern with another compatible one, such as a subor superclass. This is done through a clickable graphic visualization of the ontology neighbourhood of the currently selected class.
The Multi-Facet search portals entioned earlier can also be thought of as user interfaces for creating a very constrained subset of complex graph patterns. While in the simple case the query is formulated as searching for an information with particular properties (place of use, material, etc.), in a wider sense the definitions of how the objects map to the views can be arbitrarily complex and involve graph navigation, as for example where museum items are not directly annotated to particular event types, but the link is drawn from a combination of item type and material, for example. The portals based on OntoViews also provide limited support for a statistical view to the data, because they can group the result set according to a selected category tree or other grouping definition. This provides the portals with the capability to answer questions such as: “From which parts of the world do toys used in 18th century Finland come from”. In complex queries where the selection is based on a global intersection of distinct selectors, the individual constraints need not be ontological. In OntoViews -based portals, categories and items can be filtered using keyword constraints, while [30] contains a method that allows one to treat keyword search terms as ontological classes whose instances have fuzzy membership values. A fuzzy logic formalism is then used to calculate relevance with respect to the entire query pattern formalized as a fuzzy logic statement.
Those are the five of directions in the Semantic Searching and many of they have implemented and some of they are under developing. Although we have to first need sufficient data for these searching. Still there aren’t enough resources in the Semantic Web.
Problem Solving
Describing a problem and searching for a solution by inferring one based on ontological knowledge is one of the core use cases often associated with the vision of the semantic web. However, actual implementations of such are rare and they are usually quite
simple. [31] describes a query language for the semantic web, which, despite mostly being intended for simpler SQL-like queries is based on a DL-reasoner, and allows for a form of “if-then” queries. This functionality in turn has been used in the Wine Agent demonstration portal2. Here, the user enters information on the flavors in a dish, and the system infers from the ontological knowledge a recommendation for a wine suitable to complement those flavors.
Connecting Path Discovery
While usually property relations are used to traverse from an interesting resource to another, sometimes what is interesting are actually the paths in the graph connecting the items. In the realized vision for the semantic web, a huge amount of varied semantic data will be available to be mined for semantic connections. An example of a domain where this could prove very useful is the national security domain, where there is a need for locating connections and patterns suggesting possible security threats, such as emerging links between known terrorists and potential recruits. A major problem here is how to define a link interestingness measure in a way which both eliminates uninteresting relations (“Company A and terrorist organization B are related because they both operate in the same country”) but is still general enough to be of use in locating complex, hidden relationships in the data. One attempt at formulating an easily calculable general purpose requirement for interesting associations is described in [32].
05) Application of Semantic Searching
Hear I present some of builded and under developing Semantic Search engines and it gives us a good feeling that is its not too far to feel the Semantic Web experiences to the human beings.
· Swoogle
Swoogle is a search engine for Semantic Web documents, terms and data found on the Web. Swoogle employs a system of crawlers to discover RDF documents and HTML documents with embedded RDF content. Swoogle reasons about these documents and their constituent parts (e.g., terms and triples) and records and indexes meaningful metadata about them in its database.
Swoogle provides services to human users through a browser interface and to software agents via web services. Several techniques are used to rank query results inspired by the PageRank algorithm developed at Google but adapted to the semantics and use patterns found in semantic web documents.
Swoogle was developed at and is hosted by the University of Maryland, Baltimore County (UMBC) with funding from the US DARPA and National Science Foundation agencies [33].
· Hakia
hakia is an Internet search engine. The company has invented QDEXing technology, an alternative new infrastructure to indexing that uses SemanticRank algorithm, a solution mix from the disciplines of ontological semantics, fuzzy logic, computational linguistics, and mathematics. Founded in 2004, the company is privately held and based in New York City.hakia was founded by Riza Berkan, a nuclear scientist by training with a specialization in artificial intelligence and fuzzy logic, and Pentti Kouri, a New York-based economist and venture capitalist. Professor Victor Raskin, a father of ontological semantics and noted international authority in the field of computational linguistics, serves as hakia’s scientific advisor.
Members of its board include former Senator Bill Bradley, Pentti Kouri, Riza C. Berkan, Ryszard Krauze, Anuj Mathur, Murat Vargi and John Grzymala. hakia has raised $21 million from private equity investors [34].
· Shoe
SHOE is a small extension to HTML which allows web page authors to annotate their web documents with machine-readable knowledge. SHOE makes real intelligent agent software on the web possible.
HTML was never meant for computer consumption; its function is for displaying data for humans to read. The “knowledge” on a web page is in a human-readable language (usually English), laid out with tables and graphics and frames in ways that we as humans comprehend visually.
Unfortunately, intelligent agents aren’t human. Even with state-of-the-art natural language technology, getting a computer to read and understand web documents is very difficult. This makes it very difficult to create an intelligent agent that can wander the web on its own, reading and comprehending web pages as it goes.
SHOE eliminates this problem by making it possible for web pages to include knowledge that intelligent agents can actually read [35].
06) Future Direction of Semantic Searching
Still these Semantic Searching is under developing area. So basically future of this will have so many improvements. One of the main developing in future is how to improve the algorithm efficiency of the algorithms when deal with objects in Semantic Web and they will have to deal with so many data load to grab the appropriate things. Another main thing is in Semantic Searching, it would have to understand the patterns which are used by the human beings used to derive the searching resources. According to those directions Semantic Searching will be drove.
07) Discussion
There are many common patterns found in the approaches described in this review paper. On the technique level, it would seem that in the context of working within an RDF model, common methodologies utilized, separatable modular steps, which could quite feasibly be made use of in most of the systems, regardless of research direction. But not only are the methodologies general, it would seem that some of the research directions can be combined. Simple concept location can be seen as a precursor and subset of the interfaces allowing selection by more complex graph patterns. Fuzzy logic formalisms and fuzzy concepts allow for the combining keyword search results as equal partners in complex constraint querying. And while usually complex constraint queries have focused on models where individuals and classes are the interesting information items, also the relations are present as equal partners in all the graph pattern, path and logic formalisms. And after finding a result set using complex constraints, there is no reason not to apply the graph traversal algorithms to locate additional result items. The only direction that does not neatly wrap into the others in this way is inferencebased problem solving. While it can always be said that any search problem is indeed a problem to be solved, the leap here would be much longer. While for example the OWL-QL query language[31] is based on reasoning, reasoners and inference seem in general a much heavier tool than need be for the most usual cases of semantic search.
In application of semantic web I only consider some of the semantic search engines only. Currently there are so many semantic search engines are developed and under developing.
Acknowledgement
My very special thanks goes to my supervisor Mrs. Champika Manel for helping me throughout my independent study by advising, directing and encouraging me to achieve success in my literature survey and also I would like to thanks every one of the faculty, my friends who helped me so much to write this review paper.
References
[1] http://en.wikipedia.org/wiki/Web_3.0
[2] http://en.wikipedia.org/wiki/Semantic_search
[3] http://www2003.org/cdrom/papers/refereed/p7 79/ess.html
[4] www.seochat.com/c/a/Search-Engine-Optimization-Help/Search-Engines-and-Algorithms-Semantic-Search/
[5] http://www.w3.org/RDF/ [6] http://www.w3.org/TR/rdf-concepts/
[7] http://www.w3schools.com/semweb/default.asp
[8] http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6VRG-48CFM8P-2&_user=10&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_version=1&_urlVersion=0&_userid=10&md5=1b560230040604cb75861b2a0ec5877f
[9] http://www.websemanticsjournal.org/papers/20040701/document5.pdf
[10] http://www.w3.org/2002/05/tap/
[11] Moldovan, D.I., Mihalcea, R.: Using wordnet and lexical operators to improve internet searches. IEEE Internet Computing 4 (2000) 34–43
[12] Buscaldi, D., Rosso, P., Arnal, E.S.: A wordnet-based query expansion method for geographical information retrieval. In: Working Notes for the CLEF Workshop. (2005)
[13] Kruse, P.M., Naujoks, A., Roesner, D., Kunze, M.: Clever search: A wordnet based wrapper for internet search engines. In: Proceedings of the 2nd GermaNet Workshop. (2005)
[14] Guha, R., McCool, R., Miller, E.: Semantic search. In: WWW ’03: Proceedings of the 12th international conference on World Wide Web, ACM Press (2003) 700–709
[15] Rocha, C., Schwabe, D., de Arag˜ao, M.P.: A hybrid approach for searching in the semantic web. In: Proceedings of the 13th international conference on World Wide Web. (2004) 374–
383
[16] Airio, E., J¨arvelin, K., Saatsi, P., Kek¨al¨ainen, J., Suomela, S.: Ciri - an ontology-based query interface for text retrieval. In Hyv¨onen, E., Kauppinen, T., Salminen, M., Viljanen, K. Ala-Siuru, P., eds.: Web Intelligence: Proceedings of the 11th Finnish Artificial Intelligence Conference. (2004)
[17] Heflin, J., Hendler, J.: Searching the web with shoe (2000)
[18] Maedche, A., Staab, S., Stojanovic, N., Studer, R., Sure, Y.: Seal - a framework for developing semantic web portals. In: Advances in Databases, Proceedings of the 18th British National Conference on Databases. (2001) 1–22
[19] M¨akel¨a, E., Hyv¨onen, E., Sidoroff, T.: View-based user interfaces for information retrieval on the semantic web. In: Proceedings of the ISWC-2005Workshop End User SemanticWeb Interaction. (2005)
[20] M¨akel¨a, E., Hyv¨onen, E., Saarela, S., Viljanen, K.: OntoViews - A Tool for Creating Semantic Web Portals. In: Proceedings of the Third Internation Semantic Web Conference, Springer Verlag (2004)
[21] Reynolds, D., Shabajee, P., Cayzer, S.: Semantic Information Portals. In: Proceedings of the 13th International World Wide Web Conference on Alternate track papers & posters, ACM Press (2004)
[22] Hyv¨onen, E., M¨akel¨a, E., Salminen, M., Valo, A., Viljanen, K., Saarela, S., Junnila, M., Kettula, S.: Museumfinland – finnish museums on the semantic web. Web Semantics: Science, Services and Agents on the World Wide Web 3 (2005) 224–241
[23] Hyv¨onen, E., M¨akel¨a, E.: Semantic autocompletion. (2005) [24] M¨akel¨a, E., Viljanen, K., Lindgren, P., Laukkanen, M., Hyv¨onen, E.: Semantic yellow page service discovery: The veturi portal. In: Poster paper, 4th International Semantic Web Conference. (2005)
[25] Karger, D.R., Bakshi, K., Huynh, D., Quan, D., Sinha, V.: Haystack: A general-purpose information management tool for end users based on semistructured data. In: Proceedings of the CIDR Conference. (2005) 13–26
[26] Quan, D., Huynh, D., Karger, D.R.: Haystack: A platform for authoring end user semantic web applications. In: Proceedings of the Second International Semantic Web Conference. (2003) 738–753
[27] Teevan, J., Alvarado, C., Ackerman, M.S., Karger, D.R.: The perfect search engine is not enough: a study of orienteering behavior in directed search. In: Proceedings of the Conference on Human Factors in Computing Systems, CHI. (2004) 415–422
[28] Athanasis, N., Christophides, V., Kotzinos, D.: Generating on the fly queries for the semantic web: The ics-forth graphical rql interface (grql). In: Proceedings of the Third International Semantic Web Conference. (2004) 486–501
[29] Catarci, T., Dongilli, P., Mascio, T.D., Franconi, E., Santucci, G., Tessaris, S.: An ontology based visual tool for query formulation support. In: Proceedings of the 16th Eureopean Conference on Artificial Intelligence, IOS Press (2004) 308–312
[30] Zhang, L., Yu, Y., Zhou, J., Lin, C., Yang, Y.: An enhanced model for searching in semantic portals. In: WWW ’05: Proceedings of the 14th international conference on World Wide Web, New York, NY, USA, ACM Press (2005) 453–462
[31] Fikes, R., Hayes, P., Horrocks, I.: Owl-ql: A language for deductive query answering on the semantic web. Technical report, Knowledge Systems Laboratory, Stanford University, Stanford, CA (2003)
[32] Anyanwu, K., Sheth, A.P.: r-queries: enabling querying for semantic associations on the semantic web. In: Proceedings of the 12th international conference on World Wide Web. (2003) 690–699
[33] http://swoogle.umbc.edu/index.php?option=com_swoogle_manual&manual=search_overview
[34] http://company.hakia.com/
[35] http://www.cs.umd.edu/projects/plus/SHOE/
