KR20090072680A - Personalized semantic based web document recommendation system and its method - Google Patents

Abstract

The present invention relates to a personalized semantic based web document recommendation system and method thereof. The present invention includes a document collection step of collecting at least one web (web) document using a web robot; A generalization ontology generation step of generating a generalization ontology for the collected at least one web document; A monitoring step of monitoring a web document visited by a user; An inference step of inferring a concept related to a web document visited by the user (hereinafter referred to as "user preference concept") from the generalization ontology; A preference information ontology generation step of generating a preference information ontology for the user, to which a correlation between the inferred user preference concept and the generalized ontology is given; A receiving step of receiving a document recommendation request from the user; And a recommendation step of recommending at least one web document from the generalization ontology to the user in consideration of the user preference concept and document similarity according to the document recommendation request. Provides a personalized already based web document system.

Description

System for Recommending Personalized Meaning-Based Web-Document And Its Method}

The present invention relates to the field of web document recommendation, and more particularly, to generate a generalized ontology defining semantic relations from documents collected by a web robot in order to select documents requested by a user, and to generate a generalized ontology from a document visited by a user. Personalized semantic based web document that infers association with concepts, creates preference information ontology that reflects user's preference concept and preference, and recommends web document that meets user's intention by using generalization ontology and preference information ontology. A recommendation system and method thereof.

In the field of web search, the user spends a lot of time searching for the desired information. Various web document recommendation systems have been developed for efficient web search.

The web document recommendation system according to the related art is a method of recommending a user's preferred web document by measuring similarity using a frequency of words or a probabilistic method based on a text mining technique. Despite the nesting, if the same word does not appear in the document, it was not recommended.

In particular, a web document recommendation system that does not take into account semantic relations is possible only in cross-sectional recommendation. For example, when a user is interested in "baseball", only documents containing the word "baseball" may be recommended. Considering the semantic relationship, information about semantic related concepts such as "catcher" and "pitcher" related to "baseball" can be recommended.

In addition, with the recent research on the semantic web, a semantic web-based search engine called a semantic portal is being developed. However, it is difficult to give a connection with the ontology expressing a conceptual relationship with a document. This causes the problem of contradicting the result of wanting to recommend various information through conceptual association by accurately identifying the semantic information required by the user in the field of web document recommendation.

The present invention solves the above problems and is proposed according to recent trends and requests, and generates a generalized ontology that defines semantic relationships from documents collected by a web robot to select documents requested by the user. Inferring associations with the concepts of generalized ontology from the visited documents, creating preference information ontology reflecting user's preference concept and preference, and recommending web documents that meet the user's intention by using generalization ontology and preference information ontology. Its purpose is to provide a personalized semantic based web document recommendation system and method thereof.

As an aspect of the present invention for achieving the above object, the personalized meaning-based web document recommendation method according to the present invention, document collection step of collecting at least one web (web) using a web robot; A generalization ontology generation step of generating a generalization ontology for the collected at least one web document; A monitoring step of monitoring a web document visited by a user; An inference step of inferring a concept related to a web document visited by the user (hereinafter referred to as "user preference concept") from the generalization ontology; A preference information ontology generation step of generating a preference information ontology for the user, to which a correlation between the inferred user preference concept and the generalized ontology is given; A receiving step of receiving a document recommendation request from the user; And a recommendation step of recommending at least one web document to the user from the generalized ontology in consideration of the user preference concept and document similarity according to the document recommendation request.

The generalized ontology generation step may be characterized by defining each instance for each of the at least one web document collected by the web robot. Here, the instance may define an identifier for identifying the web document as an instance name and include information regarding at least one of an annotation, an image, a title, and a URL describing the web document.

The inferring step may be characterized by indexing all instances of the generalized ontology into a vector model and inferring the concept to which the document most similar to the web document visited by the user belongs as the user preference concept.

The preference information ontology generation step may include generating, as at least one instance of the preference information ontology, at least one concept most similar to the user preference concept among concepts included in the generalized ontology, and generating the preference information ontology. It may be a technical feature to assign an association to the generated at least one instance.

The preference information ontology generation step may generate preference information indicating the user's preference for the created instance, wherein the preference is determined in relation to the frequency of visit of the user to the web document associated with the generated instance. It can be made into a technical feature.

The receiving step may be performed by receiving a specific query from the user or receiving a selection signal for a specific web document.

The recommendation step may include a similarity between instances of the specific query or the specific web document and the generalization ontology (hereinafter referred to as "first similarity"), a concept of the specific query or the specific web document and instances of the generalization ontology. The technical feature may include recommending the at least one web document in consideration of the similarity between concepts (hereinafter referred to as “second similarity”) and the user preference.

The first similarity may be technically measured by applying a cosine similarity measuring method to the specific query word or the specific web document and the instances of the generalized ontology. And wherein the second similarity includes a first set consisting of at least one concept inferred with respect to the particular query or the particular web document and a second set consisting of concepts of instances of the generalization ontology. The technical feature may be measured by expressing the similarity between the first set and the second set.

In another aspect of the present invention for achieving the above object, the personalized semantic based web document recommendation system according to the present invention collects at least one web document using a web robot and the collected at least one web document. A web document collection unit generating a generalization ontology for the web document; Monitor the web document visited by the user, infer a concept related to the web document visited by the user (hereinafter referred to as "user preference concept") from the generalized ontology, and relate the inferred user preference concept to the generalized ontology A preference information generator configured to generate a preference information ontology for the user to which the user is assigned; And a document recommendation unit for recommending at least one web document from the generalized ontology to the user in consideration of the user preference concept and document similarity according to the document recommendation request received from the user.

According to the personalized semantic based web document recommendation system and method according to the present invention, a generalized ontology is defined by defining semantic relationships from documents collected by a web robot, and the user's preference concept and preference are grasped from documents visited by the user. By generating a preference information ontology, using the generalization ontology and the preference information ontology can recommend a personalized web document that accurately reflects the user's intention.

The above objects, features and advantages of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Like numbers refer to like elements throughout. In addition, when it is determined that the detailed description of the known function or configuration related to the present invention may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

1 is a block diagram of a personalized semantic based web document recommendation system according to an embodiment of the present invention. 2 is a flowchart of a personalized semantic based web document recommendation method according to an embodiment of the present invention. 1, 2 and the necessary drawings, a personalized semantic based web document recommendation system and method thereof according to an embodiment of the present invention will be described in detail.

As illustrated in FIG. 1, the personalized semantic based web document recommendation system 100 may include a web document collection unit 20, a preference information generation unit 40, and a document recommendation unit 50. . The functions of the web document collection unit 20, the preference information generation unit 40, and the document recommendation unit 50 will be described briefly as follows.

The web document collection unit 20 collects at least one web document using a web robot and generates a general ontology 24 related to the collected at least one web document.

The preference information generation unit 40 infers the concept of the web document visited by the logged-in user from the generalization ontology 24, and uses the inferred concept as the user preference concept of the user. Generate 45.

The document recommendation unit 50 recommends at least one web document to the user using the general ontology 24 and the preference information ontology 45 according to a document recommendation request received from the user.

Hereinafter, a detailed operation of the personalized semantic based web document recommendation system 100 and a personalized semantic based web document recommendation method according to an embodiment of the present invention will be described in detail.

The web document collection unit 20 collects at least one web document using a web robot [S101]. The web robot can collect all documents hyperlinked from the first given Seed URL 10 as a starting point. In this case, the web document collection unit 20 may use a queue 21. A queue is a queue. The web document collecting unit 20 stores the initially given Seed URL 10 in the queue 21 and collects a document corresponding to the Seed URL 10 stored in the queue 21 [S101] All URLs included in the document corresponding to the Seed URL 10 are obtained [S103] and added to the queue 21 [S105]. In addition, the web document collection unit 20 may collect documents corresponding to URLs stored in the queue 21 [S101]. In this manner, the web document collection unit 20 may repeat the steps S101 to S105 a predetermined number of times.

The web document collection unit 20 generates a generalization ontology 24 for the collected at least one web document [S107]. The generalized ontology 24 grasps the subject of the collected at least one web document and defines conceptual connectivity.

3 is a diagram illustrating an example of generating a generalized ontology. In general, an ontology for a web document is composed of information about concepts, instances, and relationships among the concepts included in the web document. The web document collection unit 20 may generate the generalization ontology 24 in a manner of defining different instances for each of the collected at least one web document. Each of the instances defines an identifier 60 for identifying the web document as an instance name, an annotation 61, an image 62, and a title 63 describing the web document. And at least one of URL 64.

The preference information generation unit 40 monitors the web document 41 visited by the user who has logged in to the system 100 [S109], and the concept related to the web document 41 visited by the user (hereinafter referred to as "user"). Preference concept ”is deduced from the generalization ontology 24 using the ontology inference engine 46 [S111]. Here, the preference information generator 40 indexes all instances of the generalized ontology 24 as a vector model, and recognizes the concept that the document 43 most similar to the web document 41 visited by the user belongs to the user preference. It can be inferred from the concept.

4 is a diagram illustrating an example of step S111. In step S111, in order to predict what kind of document the user is interested in, inferring which concept the web document 41 visited by the user belongs to. As shown in FIG. 4, the preference information generation unit 40 indexes the annotations 61 of all instances included in the generalization ontology 24 into the vector model 70, and the ontology inference engine 46. ), The concept of the document 43 most similar to the web document 41 visited by the user can be inferred as the concept of the visited web document 41 with respect to the indexed instances of the generalized ontology 24. have. In FIG. 4, the Concepts function 71, which infers the concept of a document, receives a query q, measures cosine similarity with the document d indexed in the vector model, and returns a set of concepts to which the document having the highest similarity belongs.

A vector model is a model that expresses a document on a vector to grasp the association of the documents. In order to represent a document on a vector, all words appearing in the document can be expressed in terms of TF (Term Frequence) -IDF (Inverse Document Frequence). For example, a document A is represented by "a b c c e" and a document B is represented by "a b c c e". Documents A and B may be expressed in five dimensions. Considering TF only, document A is (a, b, c, d, e) = (1, 1, 2, 1, 0), and document B is (a, b, c, d, e) = (1, 1, 2, 0, 1). The IDF lowers the weight for words that appear too frequently, taking into account document frequency. For example, the word a appears in both document A and document B, so it is not a good word to distinguish between documents. Therefore, the weight of the word a is to be low and to have a high weight for the word d or e. In this way, the expression on the vector for all documents is called an index.

A cosine measurement method may be used to measure the similarity between documents represented in the vector model. As shown by reference numeral 71 in FIG. 4, a cosine measurement method is a method of normalizing two documents to a unit vector and measuring a cosine angle between two documents. Cosine measurement is a method commonly used in vector models of information retrieval.

In addition, the preference information generation unit 40 generates a preference information ontology 45 about the user, to which the association of the inferred user preference concept and the generalized ontology 24 is given (S113). 5 is a diagram illustrating an example of generating a preference information ontology 45.

In step S113, the at least one concept most similar to the user preference concept among the concepts included in the generalization ontology 24 is generated as at least one instance of the preference information ontology 45, and the at least one most similar. It is possible to give an association to the concept of the generated at least one instance. For example, in step S113, as shown in FIG. 5, the concept 80 to which the most similar document 43 belongs is generated as an instance 81 of the preference information ontology 45, and the most similar The document 43 and the instance 81 may be associated with a "referTo" ObjectProperty.

5 shows the relationship between the generalized ontology 24 and the preference information ontology 45. For example, when a user is interested in the concept of "Machine Learning" of the generalization ontology 24, a "Machine Learning" instance is created in the preference information ontology 24, as shown in FIG. As a source of preference for this concept, a "referTo" relationship is formed. In the example of FIG. 5, the background in which the user prefers the concept of "Machine Learning" is that a "referTo" relationship is formed because the user previously visited a document suitable for an instance called "Bayesian Network".

In step S113, preference information indicating the preference of the user with respect to the generated instance 81 may be generated. The user preferences indicate how much a user prefers the created instance 81. The user's preference may be determined in relation to the frequency of visits of the user to the web document associated with the created instance 81. Referring to FIG. 5, the user's preference may be expressed by having a “hasWeight” relationship. That is, in the preference information ontology 45, the concepts preferred by the user are represented by instances, and the user may know which similar web documents have been visited previously through the "referTo" relationship, and the user has the "hasWeight" relationship. You will know how much you prefer a particular concept or instance.

6 is a diagram illustrating an example of a method of measuring a user's preference with respect to the generated instance 81. As illustrated in FIG. 6, the preference information generation unit 40 uses instances of the preference information ontology 45 as a key and a weight value corresponding to a key having a weight value of the “hasWeight” DatatypeProperty associated with the instance. User preferences can be measured or modified by using preference dictions and inferred concepts of documents currently visited by users. The dictionary includes conceptual user preference information.

The preference information generator 40 determines whether the generated key set (PreferenceDic.KeySet) includes all the concepts inferred from the document visited by the user. If it does not include all the inferred concepts, the concept (c) is initialized to 0 only for concepts (NewConcepts) that are not included in the dictionary (PreferenceDic [c] = 0). The initialization process is intended to specify that the previous preferences of the concepts are zero. The weight of the currently visited document is determined by the "Occur" function. If the concept of the document visited by the current user matches the concept of PreferenceDic, a value of 1 may be returned.

The preference information generation unit 40 may update the preference with respect to all the concept keys of the added dictionary in consideration of the concept preference 83 of the previous preference and the document visited by the user (82). For example, when a user first visits a web document having a concept of "network", PreferenceDic [network] = 1 because the concept is not defined in the preference information ontology 45. If the user first visited a web document with the concept of "database", then PreferenceDic [network] = 0.5 and PreferenceDic [database] = 1 (σ = 0.5). By the sigma value, the degree to which the value of the previous preference becomes smaller can be set. When the user visits "network" again, PreferenceDic [network] = 1.26, PreferenceDic [database] = 0.5. Thus, the preference for documents that are not frequently visited becomes gradually smaller and closer to zero. As a result, a high preference may be generated for a concept that a user is interested in recently or a concept that is frequently interested.

The document recommendation unit 50 receives a document recommendation request from the user [S115]. The document recommendation request may be made by the user inputting a specific query or by the user selecting a specific web document. For example, when the specific web document is selected by the user, the user visits the specific web document.

The document recommender 50 may infer the specific query word or similar concept and document from the preference information ontology 45 using the ontology inference engine 46 [S117]. The ontology inference engine 46 may perform hierarchical inference of concepts and new concept inference of instances. Examples of the ontology inference engine 46 include FACT ++, Pellet, Kaon2, RacerPro, and the like.

For example, in the case of vertical relation inference, if the concept of "pitcher" is defined as "thrower" and "baseball", the ontology inference engine 46 is a sub-concept of "thrower" who throws the ball. Infer that baseball is a subconcept.

As an example of a new conceptual inference of an instance, when there is "Lee Seung-yeop" as an instance of the concept of "baseball", the ontology inference engine 46 infers a conceptual relationship so that "Lee Seung-yeop" is a sub-concept of the "baseball" concept. Infer that it is an instance of "the other."

The document recommendation unit 50 infers the upper and lower relations of the concepts inferred from the generalized ontology 24 using the ontology inference engine 46 and generates a document similar to the query word input by the user based on the changed upper and lower relations. The concept can be inferred from the concept of a query.

In response to the document recommendation request, the document recommendation unit 50 recommends at least one web document from the general ontology 24 to the user in consideration of the user preference concept and document similarity [S119]. The document recommender 50 may recommend a document suitable for a user's intention in consideration of the similarity of the vector model, the weight of the user preference concept, and the concept weight of the user input query.

7 is a diagram illustrating an example of an equation for performing step S119. As shown in FIG. 7, the document recommendation unit 50 has a similarity 85 of a vector model, a similarity 86 between a concept of a query and a concept of a document, for the indexed instances of the generalized ontology 24. The user may recommend at least one web document by calculating a score considering the weight 82 of the user preference concept and sorting the score in descending or ascending order.

The similarity 85 of the vector model means the similarity between the specific query word or the specific web document and instances of the generalization ontology 24. The similarity 85 of the vector model may be measured by applying a cosine similarity measuring scheme to the specific query word or the instance of the specific web document and the generalization ontology 24. That is, the similarity 85 of the vector model may be calculated by calculating the cosine similarity between the query input by the user and the indexed document in the indexed vector model.

Similarity 86 between the concept of the query and the concept of the document means the similarity between the concept of the particular query or the particular web document and the concepts of instances of the generalization ontology 24. The similarity 86 between the concept of the query and the concept of the document is a first set consisting of at least one concept inferred for the particular query or the particular web document and a concept consisting of the concepts of instances of the generalization ontology 24. In a concept set including two sets, it may be measured by expressing the similarity between the first set and the second set. For example, the first set, which is a conceptual set of query terms, is deduced as {'baseball', 'catcher'}, and the second set, which is a concept set of a comparison target document, is {'baseball', 'pitcher'}. The concept set including the first set and the second set is {'baseball', 'catcher', 'pitcher'}. Therefore, the similarity 86 between the concept of the query and the concept of the document is 1/3. Also, for example, if the first set is {'baseball', 'catcher'} and the second set is {'baseball', 'catcher'}, the similarity between the concept of the query and the concept of the document 86 Becomes 2/2. In other words, the similarity 86 between the concept of the query and the concept of the document indicates how overlapping the concept of the query and the document is.

8 is a diagram illustrating an example of a document recommended by step S119. In the interface 90 that collects the documents visited by the user and provides the recommendation information, "network" is entered as the recommendation keyword. The concept of the recommended document is shown on the right side of the screen 92, and the documents are listed on the left side of the screen in consideration of their priority. When a user requests a document recommendation on "network", as shown in FIG. 8, a personalized semantic based document may be recommended by weighting a concept preferred by the user so far.

The personalized meaning-based web document recommendation method according to the present invention described above may be provided by recording on a computer-readable recording medium as a program for executing on a computer.

The personalized semantic based web document recommendation method according to the present invention can be executed through software. When implemented in software, the constituent means of the present invention are code segments that perform the necessary work. The program or code segments may be stored on a processor readable medium or transmitted by a computer data signal coupled with a carrier on a transmission medium or network.

Computer-readable recording media include all kinds of recording devices that store data that can be read by a computer system. Examples of computer-readable recording devices include ROM, RAM, CD-ROM, DVD ± ROM, DVD-RAM, magnetic tape, floppy disk, hard disk, optical data storage device, and the like. The computer readable recording medium can also be distributed over network coupled computer devices so that the computer readable code is stored and executed in a distributed fashion.

The present invention described above is capable of various substitutions, modifications, and changes without departing from the spirit of the present invention for those skilled in the art to which the present invention pertains. It is not limited by the drawings.

1 is a block diagram of a personalized semantic based web document recommendation system according to an embodiment of the present invention.

2 is a flowchart of a personalized semantic based web document recommendation method according to an embodiment of the present invention.

3 is a diagram illustrating an example of generating a generalized ontology.

4 is a diagram illustrating an example of step S111.

5 is a diagram illustrating an example of generating a preference information ontology 45.

6 is a diagram illustrating an example of a method of measuring a user's preference with respect to the generated instance 81.

7 is a diagram illustrating an example of an equation for performing step S119.

8 is a diagram illustrating an example of a document recommended by step S119.

<Explanation of symbols for the main parts of the drawings>

100: Web document recommendation system 10: Seed URL

20: Web document collection unit 21: Queue

24: generalized ontology 40: preference information generation unit

45: Preference Ontology 46: Inference Engine

50: document recommendation

Claims (21)

A document collection step of collecting at least one web document using a web robot; A generalization ontology generation step of generating a generalization ontology for the collected at least one web document; A monitoring step of monitoring a web document visited by a user; An inference step of inferring a concept related to a web document visited by the user (hereinafter referred to as "user preference concept") from the generalization ontology; A preference information ontology generation step of generating a preference information ontology for the user, to which a correlation between the inferred user preference concept and the generalized ontology is given; A receiving step of receiving a document recommendation request from the user; And A recommendation step of recommending at least one web document to the user from the generalized ontology in consideration of the user preference concept and document similarity according to the document recommendation request How to recommend a personalized semantic-based web document that includes. The method of claim 1, wherein the generalized ontology generation step, Personalized meaning-based web document recommendation method, characterized in that each of the at least one web document collected by the web robot defined as an instance. The method of claim 2, wherein the instance is Personalized semantic based web document recommendation characterized by defining an identifier for identifying the web document as an instance name and including information on at least one of a comment, an image, a title, and a URL describing the web document. Way. The method of claim 2, wherein the reasoning step, And indexing all instances of the generalized ontology into a vector model and inferring the concept to which the document most similar to the web document visited by the user belongs as the user preference concept. The method of claim 4, wherein the preference information ontology generation step, Generating at least one instance of the preference information ontology, the at least one concept most similar to the user preference concept among the concepts included in the generalized ontology, and generating the at least one concept and the generated at least one instance. Personalized semantic based web document recommendation method characterized by granting an association. The method of claim 5, wherein the preference information ontology generation step, Generating preference information indicating the user's preference for the created instance, And wherein said preference is determined in relation to a frequency of said user's visit to a web document associated with said generated instance. The method of claim 6, wherein the receiving step, Personalized semantic based web document recommendation method, characterized in that by receiving a specific query from the user or a selection signal for a specific web document. The method of claim 6, wherein the recommendation step, The similarity between the specific query or the instance of the specific web document and the generalization ontology (hereinafter referred to as "first similarity"), and the similarity between the concept of the specific query or the specific web document and the concepts of the instances of the generalization ontology (hereinafter referred to as "first similarity"). Personalized semantic based web document recommendation method, wherein the at least one web document is recommended in consideration of the &quot; second similarity &quot; The method of claim 8, wherein the first similarity degree, Personalized semantic based web document recommendation method characterized in that it is measured by applying a cosine similarity measurement method for the specific query or the instance of the specific web document and the generalized ontology. The method of claim 8, wherein the second similarity degree, A concept set comprising a first set of at least one concept inferred for the particular query or the particular web document and a second set of concepts of instances of the generalization ontology, wherein the first set and the second set of A personalized semantic based web document recommendation method characterized by measuring similarity. A computer-readable recording medium having recorded thereon a computer program capable of executing the method of any one of claims 1 to 10. A web document collection unit which collects at least one web document using a web robot and generates a generalization ontology for the collected at least one web document; Monitor the web document visited by the user, infer a concept related to the web document visited by the user (hereinafter referred to as "user preference concept") from the generalized ontology, and relate the inferred user preference concept to the generalized ontology A preference information generator configured to generate a preference information ontology for the user to which the user is assigned; And In response to a document recommendation request received from the user, a document recommendation unit recommending at least one web document from the generalized ontology to the user in consideration of the user preference concept and document similarity. Personalized semantic based web document recommendation system that includes. The method of claim 12, wherein the web document collection unit, Personalized meaning-based web document recommendation system, characterized in that each of the at least one web document collected by the web robot defined as an instance. The method of claim 13, wherein the instance is Personalized semantic based web document recommendation characterized by defining an identifier for identifying the web document as an instance name and including information on at least one of a comment, an image, a title, and a URL describing the web document. system. The method of claim 13, wherein the preference information generation unit, Personalized semantic based web document recommendation system, comprising: indexing all instances of the generalized ontology into a vector model and inferring the concept to which the document most similar to the web document visited by the user belongs as the user preference concept. The method of claim 15, wherein the preference information generation unit, Generating at least one instance of the preference information ontology, the at least one concept most similar to the user preference concept among the concepts included in the generalized ontology, and generating the at least one concept and the generated at least one instance. Personalized semantic based web document recommendation system characterized by assigning relevance. The method of claim 16, wherein the preference information generation unit, Generating preference information indicating a preference of the user with respect to the created instance, The preference is personalized semantic based web document recommendation system, characterized in that it is determined in relation to the frequency of visits of the user to the web document associated with the generated instance. The method of claim 17, wherein the request for document recommendation, Personalized semantic based web document recommendation system, characterized in that by receiving a specific query from the user or a selection signal for a specific web document. The method of claim 17, wherein the document recommendation unit, The similarity between the specific query or the instance of the specific web document and the generalization ontology (hereinafter referred to as "first similarity"), and the similarity between the concept of the specific query or the specific web document and the concepts of the instances of the generalization ontology (hereinafter referred to as "first similarity"). Personalized semantic based web document recommendation system, wherein the at least one web document is recommended in consideration of the &quot; second similarity &quot; The method of claim 19, wherein the first similarity degree, Personalized semantic based web document recommendation system, characterized in that measured by applying a cosine similarity measurement method for the particular query or the instance of the specific web document and the generalized ontology. The method of claim 19, wherein the second similarity degree, A concept set comprising a first set of at least one concept inferred for the particular query or the particular web document and a second set of concepts of instances of the generalization ontology, wherein the first set and the second set of A personalized semantic based web document recommendation system characterized by measuring similarity.

Images