CN104637371A - Method for embedding knowledge ontology into game model - Google Patents

Abstract

The invention relates to the technical field of game-based virtual learning, in particular to a method for embedding a knowledge ontology into a game model. The method comprises the following steps: constructing a game-oriented knowledge ontology; constructing a game atomic challenge ontology model of the game model according to a game type and constitution elements; calculating the similarity of a specific knowledge set in the knowledge ontology and each game atomic challenge ontology, and establishing map from the knowledge set to each game atomic challenge ontology according to the similarity; selecting the game atomic challenge ontology with the highest similarity in the knowledge set as an adaptive game atomic challenge model. Unified representation is performed on domain knowledge and a game by taking the ontology as the basis, and the intrinsic relation between the two is analyzed by calculating the similarity of the ontology, so automatic adaption of the knowledge and the game is realized, and the subjective influence of an individual designer is reduced.

Description

A kind of method ontologies is embedded in game model

Technical field

The present invention relates to gameization Virtual Learning technical field, particularly relate to a kind of method ontologies be embedded in game model.

Background technology

Educational games fast development also achieves good achievement, its design field generally around core concept be: the design of game needs the participation in game and education two fields to participate.But for a long time, the game designing amusing products and the education conveyed knowledge belong to two diverse industries, adopt different mentalities of designing and construction method separately, are therefore difficult to the balance reaching " educational " and " recreational ".

In design of the education game, Waraich emphasize a significant study must effectively story background and motivation to the overall process impelling learner to complete study.Kristian proposes a kind of design of the education game framework tested based on fluid, the teaching element such as instructional objective, the content of courses joins in the stream experience of the process of game by participating in and feeding back by the method, allowing learner in gaming with problem, completing learning cognition by completing Mission Objective.

About the fusion method of knowledge and game, some scholars are from this body structure of knowledge and feature, and then the knowledge model built in game attempts knowledge to embed in game.Brunot proposes the cognitive model of a kind of knowledge based signature analysis and knowledge classification, adopts Object-oriented Technique to describe learning activities and learner's thought process, cognitive process is carried out the degree of depth with game link and is combined.Miroslav proposes a kind of based on learning object establishment education resource fragment, and then balance is intellectual and the knowledge model of game; Use unified modeling language (Unified Modeling Language, and extend markup language (Extensible Markup Language UML), XML) come design games and Knowledge Element model, meta-model concept of playing introduces research field.

A large amount of researcher has carried out a series of research for the problems referred to above, and current achievement mainly comprises the design guidance theory of macroscopic aspect, reference method or relates to the empirical model of microcosmic game element design.But in actual design operation, do not occur that the formalization method of generally acknowledging or template instruct, more do not formed with objective law or theorem, as with reference to, with the method for normalizing of universality or automated tool.Therefore in most cases, deviser or developer still need to carry out more subjectivity and create, and to exploitation, designer requires higher, should correct understanding domain knowledge intension, are familiar with essential characteristic and the method for designing of game again.And in actual design and performance history, often the deviation of knowledge understanding and scarcity easily cause " educational " or " science " to lack to fail to agree with the original intention of design of the education game; Or cause whole game " recreational " not strong because game design is lacked experience, make it more to be similar to a teaching software, the object continuing to attract participant can not be reached; Or " educational " and " recreational " forced combination, two processes making game experiencing and learning link become completely isolated, had both destroyed game overall experience and had lost certain interest, also have impact on final results of learning simultaneously.

Summary of the invention

The technical problem to be solved in the present invention there is provided a kind of method of domain knowledge and game mechanism being carried out automatic adaptation and also merging.

In order to solve the problems of the technologies described above, the invention provides a kind of method ontologies be embedded in game model, comprising the following steps:

Build the ontologies of object game;

According to type of play and constitution element, build this model of game atom challenge body;

Calculate described ontologies and atom of playing described in each challenges the similarity of body, and according to described similarity, to set up in described ontologies specific knowledge body collection to the mapping of atom challenge body of playing described in each;

Choose the described game atom the highest with described specific knowledge body collection similarity to challenge body and challenge model as the game atom with described specific knowledge body collection adaptation.

Preferably, the concrete steps that calculation knowledge body and atom of playing described in each challenge the similarity of body comprise:

In described ontologies, selected described specific knowledge body collection to be mapped;

The number of more described specific knowledge body centralized concept class and atom of playing described in each challenge the number of concept cluster in ontology model, if difference, described similar to being set to 0, otherwise carry out next step calculating;

Calculation knowledge body and atom of playing described in each challenge the comprehensive similarity of ontology model.

Preferably, described specific knowledge body collection comprises the information of the information of the first concept cluster, the information of the first concept classes relation, the information of the first event class, the information of the first event classes relation and the first concept cluster and the first event classes relation.

Preferably, described game atom challenge body comprises the information of the information of the second concept cluster, the information of the second concept classes relation, the information of second event class, the information of second event classes relation and the second concept cluster and second event classes relation.

Preferably, described comprehensive similarity obtains as follows:

Described comprehensive similarity is the product of the similarity of the similarity of the similarity of described first concept cluster and described second concept cluster, described first event class and described second event class, described first concept classes relation and described second concept classes relation, described first event classes relation and the similarity of described second event classes relation and the similarity of described first concept cluster and the first event classes relation and described second concept cluster and second event classes relation.

Preferably, described similarity calculates according to method below:

The description collection setting the first concept and the second concept is respectively A, B, then the similarity of the first concept a and the second concept b is tried to achieve by following formulae discovery:

$\mathrm{Sim}(a,b)=\frac{|A\∩B|}{|A\∩B|+\mathrm{\α}(a,b)|A/B|+(1-\mathrm{\α}(a,b))|B/A|}---\left(1\right)$

Wherein, A ∩ B represents the common factor describing collection A and B, and A/B represents the difference set describing collection A and B, and α (a, b) is the depth function of the first concept a and the second concept b, is tried to achieve by following formulae discovery:

$\mathrm{\α}(a,b)=\left\{\begin{array}{c}\frac{\mathrm{depth}\left(a\right)}{\mathrm{depth}\left(a\right)+\mathrm{depth}\left(b\right)},\mathrm{depth}\left(a\right)\≤\mathrm{depth}\left(b\right)\\ 1-\frac{\mathrm{depth}\left(a\right)}{\mathrm{depth}\left(a\right)+\mathrm{depth}\left(b\right)},\mathrm{depth}\left(a\right)>\mathrm{depth}\left(b\right)\end{array}\right.---\left(2\right)$

Wherein, 0≤α≤1, depth (a) represents the shortest path distance from the first concept a to root node; Depth (b) represents the shortest path distance from the second concept b to root node; First concept is the information of the information of the first concept cluster, the information of the first concept classes relation, the information of the first event class, the information of the first event classes relation or the first concept cluster and the first event classes relation, and corresponding described second concept is the information of the information of the second concept cluster, the information of the second concept classes relation, the information of second event class, the information of second event classes relation or the second concept cluster and second event classes relation.

Preferably, described description integrates as synset, feature set or semantic neighbours collection.

Preferably, described mapping comprises class mapping and relationship map; Described class maps and comprises: the mapping between concept cluster, the mapping between event class; Described relationship map comprises: the mapping of concept classes relation, the mapping of event classes relation, the mapping of concept cluster and event classes relation.

Technique scheme tool of the present invention has the following advantages: method ontologies be embedded in game model provided by the invention carries out unified representation to domain knowledge and game based on body, and the essence analyzed therebetween by calculating body similarity associates, and realizes the automatic adaptation of knowledge and game.Reduce the subjective impact of deviser's individuality, because the shortage of domain knowledge causes " educational " lack or lack experience due to game design and cause the problems such as " recreational " disappearance.Meanwhile, the establishment of game challenge ontology library and domain knowledge body frame achieves knowledge sharing and the shared mechanism of game frame under different application field between difference game, improves the validity of design.

Accompanying drawing explanation

Fig. 1 is the method design design schematic diagram that the embodiment of the present invention provides;

Fig. 2 is the method flow diagram that the invention process provides;

Fig. 3 is the method and technology solution schematic diagram that the embodiment of the present invention provides;

Fig. 4 is the ontologies block schematic illustration that the embodiment of the present invention provides;

Fig. 5 is the game challenge ontology model schematic diagram that the embodiment of the present invention provides;

Fig. 6 is that the Knowledge Set that the embodiment of the present invention provides challenges distortion computing block diagram with game atom;

Fig. 7 is that the Knowledge Set that the embodiment of the present invention provides challenges distortion computing method process flow diagram with game atom;

Fig. 8 is child's knowledge concepts of providing of the embodiment of the present invention and hierarchical structure schematic diagram.

Embodiment

Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples for illustration of the present invention, but are not used for limiting the scope of the invention.

Method design design ontologies be embedded in game model provided by the invention is, to be relative to each other in creation process but carry out unified quantization research from different field element (mainly comprising domain knowledge, technical ability main points, game element and rule mechanism etc.), and analyzing inner essential correlativity to realize the mapping between knowledge and game mechanism.As shown in Figure 1, the method design provided for the embodiment of the present invention conceives schematic diagram.The source of " game " is game challenge, and the fusion of therefore knowledge and game is in fact realize domain knowledge to merge with the degree of depth challenged of playing.Meanwhile, domain knowledge and game challenge more close in concept aspect, the possibility that both realizations degree of depth merges is larger and have more actual operation.Wait that it is limited for propagating knowledge content game challenge type for infinite, therefore method provided by the invention has carried out the similarity assessment of limited number of times, can obtain optimum Adaptation Options.

As shown in Figure 2, be method flow diagram that the invention process provides.The method comprises: the ontologies framework building object game; According to type of play and constitution element, build the game atom challenge ontology model of game model; Challenge the calculating of body similarity according to ontologies and game atom, realize the mapping of ontologies to game atom challenge body; Obtain challenging model with the game atom of ontologies knowledge content adaptation.

Further, as shown in Figure 3, be method and technology solution schematic diagram that the embodiment of the present invention provides.Method ontologies be embedded in game model specifically comprises:

(1) the relevant knowledge classification in analysis knowledge engineering field, builds object game and carries out the ontologies framework of formalization representation as shown in Figure 4 to knowledge content.

(2) analyze type of play and constitution element, build game general element model, and build game atom challenge disaggregated model according to the challenge mechanism of classical Mission Objective.On game atom challenge disaggregated model concept hierarchy, carry out formalization representation and build game atom challenge ontology model as shown in Figure 5.Wherein, game challenge is the source of " game ", and game challenge is made up of multiple different game atom challenge, and game atom challenge body is the formalization representation mode of game atom challenge.

(3) analysis two isomery bodies and ontologies are challenged between body with game atom and are associated in structure and essence semantically, by the similarity of both calculating, realize the automatic mapping of playing between atom challenge and ontologies framework and be embedded in game model by ontologies.

(4) in being challenged by ontologies collection all in ontologies framework and game, all game atoms challenge the calculating of ontology model comprehensive similarity, obtain the game atom suitable with each ontologies collection and challenge ontology model.According to the game atom challenge ontology model obtained, the task of structure game challenge further and then embodiment " game ".

Further, as shown in Figure 7, the Knowledge Set provided for the embodiment of the present invention challenges distortion computing method process flow diagram with game atom.Challenge the calculating of body similarity according to ontologies and game atom, realize ontologies challenges the mapping of body concrete steps to described game atom and comprise: knowledge based body frame, builds ontologies collection to be mapped; Relatively ontologies collection and game atom challenge the number of ontology model concept class (name part of speech); Calculation knowledge body collection challenges the comprehensive similarity of ontology model with game atom, is mapped and relationship map by the class that clustering procedure carries out ontologies collection and atom of playing is challenged between ontology model.Wherein, class maps and comprises: the mapping between concept cluster and the mapping between name part of speech and name part of speech, the mapping between event class and the mapping between verb class and verb class; Relationship map comprises: the mapping of concept classes relation and the mapping of noun classes relation, the mapping of event classes relation and the mapping of verb classes relation, the mapping of concept cluster and event classes relation and the mapping of relation between verb and noun.

Further, the similarity of both Tvorsky model calculating is adopted.Suppose that in body O1, in concept a and body O2, concept b has description collection, such as synset, feature set or semantic neighbours collection.The present invention is based on Chinese thesaurus structure concept a, the synset of b as description collection, and represents with A, B, i.e. A=syn (a), B=syn (b).Wherein, A ∩ B represents the common factor of A and B, and A/B represents the difference set of A and B, and the similarity of concept a and b can be calculated by following equation tries to achieve:

$\mathrm{sim}(a,b)=\frac{|A\∩B|}{|A\∩B|+\mathrm{\α}(a,b)|A/B|+(1-\mathrm{\α}(a,b))|B/A|}---\left(1\right)$

Wherein, α (a, b) is the function of two concept degree of depth, can be calculated try to achieve by following equation:

$\mathrm{\α}(a,b)=\left\{\begin{array}{c}\frac{\mathrm{depth}\left(a\right)}{\mathrm{depth}\left(a\right)+\mathrm{depth}\left(b\right)},\mathrm{depth}\left(a\right)\≤\mathrm{depth}\left(b\right)\\ 1-\frac{\mathrm{depth}\left(a\right)}{\mathrm{depth}\left(a\right)+\mathrm{depth}\left(b\right)},\mathrm{depth}\left(a\right)>\mathrm{depth}\left(b\right)\end{array}\right.---\left(2\right)$

Wherein, 0≤α≤1, depth (a) represents the shortest path distance from concept a to root node.

Therefore, ontologies collection and the comprehensive similarity computing formula challenged between ontology model of game atom as follows:

SIM(KnSetK,ChallModt)＝Num_NounClass_Match(KnSetK,ChallModt)*SIM1*SIM2*SIM3*SIM4*SIM5 (3)

Wherein, $\mathrm{Num}\_\mathrm{NounClass}\_\mathrm{Match}(\mathrm{KnSetK},\mathrm{ChallModt})$

$=\left\{\begin{array}{c}1,\mathrm{Num}\_\mathrm{NounClass}\left(\mathrm{KnS}{\mathrm{et}}_k\right)=\mathrm{Num}\_\mathrm{NounClass}\left(\mathrm{ChallMo}{d}_t\right)\\ 0,\mathrm{Num}\_\mathrm{NounClass}\left({\mathrm{KnSet}}_k\right)\≠\mathrm{Num}\_\mathrm{NounClass}\left({\mathrm{ChallMod}}_t\right)\end{array}\right.---\left(4\right)$

Further, as shown in Figure 6, the Knowledge Set provided for the embodiment of the present invention challenges distortion computing block diagram with game atom.When ontologies collection and game atom challenge the number of ontology model concept class identical time, comprehensive similarity is the product of the similarity SIM4 of relation (relation object) and the similarity SIM5 of concept cluster and event classes relation between similarity SIM2 between similarity SIM1 between concept cluster (name part of speech), event class (verb class), the similarity SIM3 of concept classes relation, event class (verb class); When the number that ontologies collection and game atom challenge ontology model concept class is different, comprehensive similarity is zero.It is considered herein that, as long as both core name part of speech numbers are different, then similarity is 0.Any one similarity of SIM1, SIM2, SIM3, SIM4 or SIM5 calculated is 0, then comprehensive similarity is 0.Therefore the calculation process calculating comprehensive similarity is optimized for, and when the result of calculation of certain factor similarity is 0, computation process terminates.

Further, said method tries to achieve the calculating of similarity between game atom challenge ontology model that an ontologies set pair answers, challenge ontology model to obtain all game atoms suitable with certain selected ontologies collection in ontologies framework, needing to calculate this ontologies collection and all game atoms in challenging of playing challenge comprehensive similarity between ontology model.Because of present stage game atom challenge Limited Number (20 ~ 30), and then ontologies collection and the game atom sequencing of challenging similarity can't affect and is embedded in game model by ontologies, therefore the embodiment of the present invention does not consider that ontologies collection and all game atoms challenge the sequencing problem of similarity between ontology model.The form that knowledge model is corresponding represents ontologies framework, and game is completed by challenge of playing, game atom challenge composition game challenge, and game atom ontology model form represents the challenge of game atom; Framework is a structure, and just as an empty table, load and just become ontologies after knowledge content, ontologies comprises multiple ontologies collection.

Further, in the process calculating comprehensive similarity, relate to setting threshold value to judge the problem whether comprehensive similarity result corresponds to actual needs, wherein, the threshold value of setting is generally mainly screened based on subjective experience.Same, because of present stage game atom challenge Limited Number (20 ~ 30), all challenge with multiple game atom the calculating that ontology model carries out comprehensive similarity for each ontologies collection in ontologies framework, and the corresponding game atom suitable with ontologies collection challenges ontology model number less (being less than 20 ~ 30), therefore set threshold value and can ignore for the impact of the screening in the comprehensive similarity calculating in the embodiment of the present invention.

The embodiment of the present invention is for the educational games system towards preschool child's cognition, choose and require all higher preschool child's Cognitive education field to interesting and intellectual, cognitive for target with children's color, figure, logic, letter, science and technology and social general knowledge, attempt design craps game form.Wherein, logic cognition is divided into the concept of figure cognitive, the cognition of color identification, pattern recognition, local and global deformation relation; Language cognition is divided into the cognition of English alphabet cognition and upper and lower case letter association; Naturally scientific and technological cognition is divided into animal cognition, the cognition that animal associates with food, living environment; Society's general knowledge cognition is divided into national concept cognitive, country and national flag, identify build, cognition that dress ornament etc. associates; Daily living article cognition is divided into the cognition to the various things in daily life.Concrete operations ontologies be embedded in game model for preschool child's Cognitive education field are as follows:

(1) set up ontologies framework according to above-mentioned perception target, be illustrated in figure 8 concept cluster and hierarchical structure in ontologies framework further, as shown in table 1 is relation and the attribute structure of concept cluster in ontologies framework.

Table 1 child ontologies concept cluster relation and attribute

(2) in being challenged by each ontologies collection in calculation knowledge body and game, each game atom challenges the comprehensive similarity between ontology model, chooses adaptive game atom challenge ontology model and then builds game challenge.Particularly, ontologies collection five yuan of formulas are defined as: KnowlegeSet={NounClass, N_Class_Re, VerbClass, V_Class_Re, N_V_Re}, NounClass, N_Class_Re, VerbClass, V_Class_Re, N_V_Re represent concept cluster, concept classes relation, event class, event classes relation and the relation between concept cluster and event class respectively.In concrete practical operation process, usual concept cluster is run after fame part of speech, and event class gets its main actions, i.e. verb class.

Such as, for representing the relation between " country " and " national flag ", its ontologies collection five yuan of formulas can be expressed as:

{ { country, national flag }, { indicate/being indicated }, { }, { }, { } }, namely comprise two name parts of speech " country " and " national flag ", its pass is " mark " or " by indicating ".This ontologies collection without event, after three be expressed as sky.

For another example, for representing the Predatory relation between " cat " and " fish ", its ontologies collection five yuan of formulas can be expressed as:

{ { cat, fish }, { predation/prey }, { }, { }, { } }, namely comprises two name parts of speech " cat " and " fish ", and its pass is " predation " or " prey ".This ontologies collection without event, after three be expressed as sky.

Relatively ontologies collection class capacity and each game atom challenge ontology model name part of speech number, find the challenge of most of intelligence class atom and its number matches.

Calculate name part of speech similarity: the challenge of intelligence class atom is semantic without constraint to concept cluster, and giving tacit consent to its similarity is 1.Therefore name part of speech relation Similarity Measure draws:

SIM(KS1,Chall-Mapping)＝0.8,

SIM(KS1,Chall-Distribution)＝0.3,

……

SIM(KS1,Chall-Action)＝0,

SIM(KS1,Chall-Stratege)＝0,

……

Namely this ontologies collection and " intelligence challenge "---" relation "---" mapping " are played, and to challenge ontology model similarity be 0.8 to atom, playing with " intelligence challenge "---" relation "---" distribution ", to challenge ontology model similarity be 0.3 to atom, and all the other action classes, policy class challenge similarity is 0.According to Similarity Measure end value, descending sort is carried out to above-mentioned all game atom challenge ontology models: map---distribution---

In advance according to practical experience value setting threshold value 0.5, show that " intelligence " is challenged " mappings " in subclass " relation " for the game atom that similarity is the highest and challenged ontology model.

(3) based on the atom challenge ontology model of above-mentioned adaptation, and obtain according to Similarity Measure challenge ontology model with all game atoms of sort of ontologies collection fit height, provide task level game and challenge recommendation.Other ontologies collection in selected ontologies framework, carries out same operation and draws all game atom challenge ontology models mapped with it, provide corresponding task level game challenge and recommend.By the game atom of all adaptations challenge ontology model integrated game challenge, challenge with other and combine as " time pressure " etc., form as intelligence development classes such as " infant cognition are seen repeatedly " is played.

Further, complete for figure matched rule, with reference to traditional " seeing game repeatedly ", is adjusted to logic association or mapping by " infant cognition is seen repeatedly ", and two the icon lines just realizing having correlativity are eliminated.Game content module comprises 1) elimination of same hue icon matches; 2) similar fitgures icon matches is eliminated; 3) animal and food icon matches thereof are eliminated; 4) entire and part icon matches is eliminated; 5) upper and lower case letter icon matches is eliminated; 6) national coupling with flag icon, eliminates.Each modular design 10 outposts of the tax office, each outpost of the tax office provides 20 10 pairs of icons with one-to-one relationship.Under application operates in IPAD environment, user carries out man-machine interaction by touch manner.

In sum, method ontologies be embedded in game model provided by the invention carries out unified representation to domain knowledge and game based on body, and the essence analyzed therebetween by calculating body similarity associates, and realizes the automatic adaptation of knowledge and game.Reduce the subjective impact of deviser's individuality, because the shortage of domain knowledge causes " educational " lack or lack experience due to game design and cause the problems such as " recreational " disappearance.Meanwhile, the establishment of game challenge ontology library and domain knowledge body frame achieves knowledge sharing and the shared mechanism of game frame under different application field between difference game, improves the validity of design.

Last it is noted that above embodiment is only in order to illustrate technical scheme of the present invention, be not intended to limit; Although with reference to previous embodiment to invention has been detailed description, those of ordinary skill in the art is to be understood that: it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein portion of techniques feature; And these amendments or replacement, do not make the essence of appropriate technical solution depart from the spirit and scope of various embodiments of the present invention technical scheme.

Claims (8)

1. ontologies is embedded into the method in game model, it is characterized in that, comprise the following steps:

Build the ontologies of object game;

According to type of play and constitution element, build this model of game atom challenge body;

Calculate described ontologies and atom of playing described in each challenges the similarity of body, and according to described similarity, to set up in described ontologies specific knowledge body collection to the mapping of atom challenge body of playing described in each;

Choose the described game atom the highest with described specific knowledge body collection similarity to challenge body and challenge model as the game atom with described specific knowledge body collection adaptation.

2. method ontologies be embedded in game model according to claim 1, is characterized in that, the concrete steps that calculation knowledge body and atom of playing described in each challenge the similarity of body comprise:

In described ontologies, selected described specific knowledge body collection to be mapped;

The number of more described specific knowledge body centralized concept class and atom of playing described in each challenge the number of concept cluster in ontology model, if difference, described similar to being set to 0, otherwise carry out next step calculating;

Calculation knowledge body and atom of playing described in each challenge the comprehensive similarity of ontology model.

3. method ontologies is embedded in game model according to claim 2, it is characterized in that, described specific knowledge body collection comprises the information of the information of the first concept cluster, the information of the first concept classes relation, the information of the first event class, the information of the first event classes relation and the first concept cluster and the first event classes relation.

4. method ontologies is embedded in game model according to claim 3, it is characterized in that, described game atom challenge body comprises the information of the information of the second concept cluster, the information of the second concept classes relation, the information of second event class, the information of second event classes relation and the second concept cluster and second event classes relation.

5. method ontologies is embedded in game model according to claim 4, it is characterized in that, described comprehensive similarity obtains as follows:

Described comprehensive similarity is the product of the similarity of the similarity of the similarity of described first concept cluster and described second concept cluster, described first event class and described second event class, described first concept classes relation and described second concept classes relation, described first event classes relation and the similarity of described second event classes relation and the similarity of described first concept cluster and the first event classes relation and described second concept cluster and second event classes relation.

6. method ontologies be embedded in game model according to claim 5, is characterized in that, described similarity calculates according to method below:

The description collection setting the first concept and the second concept is respectively A, B, then the similarity of the first concept a and the second concept b is tried to achieve by following formulae discovery:

$\mathrm{Sim}(a,b)=\frac{|A\∩B|}{|A\∩B|+\mathrm{\α}(a,b)|A/B|+(1-\mathrm{\α}(a,b))|B/A|}---\left(1\right)$

Wherein, A ∩ B represents the common factor describing collection A and B, and A/B represents the difference set describing collection A and B, and α (a, b) is the depth function of the first concept a and the second concept b, is tried to achieve by following formulae discovery:

$\mathrm{\α}(a,b)=\left\{\begin{array}{cc}\frac{\mathrm{depth}\left(a\right)}{\mathrm{depth}\left(a\right)+\mathrm{depth}\left(b\right)},& \mathrm{depth}\left(a\right)\≤\mathrm{depth}\left(b\right)\\ 1-\frac{\mathrm{depth}\left(a\right)}{\mathrm{depth}\left(a\right)+\mathrm{depth}\left(b\right)},& \mathrm{depth}\left(a\right)>\mathrm{depth}\left(b\right)\end{array}\right.---\left(2\right)$

Wherein, 0≤α≤1, depth (a) represents the shortest path distance from the first concept a to root node; Depth (b) represents the shortest path distance from the second concept b to root node; First concept is the information of the information of the first concept cluster, the information of the first concept classes relation, the information of the first event class, the information of the first event classes relation or the first concept cluster and the first event classes relation, and corresponding described second concept is the information of the information of the second concept cluster, the information of the second concept classes relation, the information of second event class, the information of second event classes relation or the second concept cluster and second event classes relation.

7. method ontologies be embedded in game model according to claim 6, is characterized in that, described description integrates as synset, feature set or semantic neighbours collection.

8. method ontologies be embedded in game model according to claim 7, is characterized in that, described mapping comprises class and maps and relationship map; Described class maps and comprises: the mapping between concept cluster, the mapping between event class; Described relationship map comprises: the mapping of concept classes relation, the mapping of event classes relation, the mapping of concept cluster and event classes relation.