A data storage method for entity data, an electronic device, and a storage medium

Abstract

The application provides a data storage method for entity data, acquires an optimal grouping result corresponding to a list of each type of entity data, and then acquires a first entity data group set, extracts features corresponding to the first entity data group to form a first entity data group identification list corresponding to the first entity data group set, stores the first entity data identification and the first entity data group corresponding to the first entity data identification into a storage domain according to the order in the first entity data group identification list, and makes similar entity data in the same entity data category in the same first entity data group through the first entity data group set obtained according to the optimal grouping result, and makes the connection between each first entity data group closer by storing the first entity data groups with high matching degrees into adjacent storage domains. Therefore, the first entity data groups with high matching degrees are stored close to each other, which facilitates subsequent extraction and query of the entity data, and improves time efficiency.

Description

Technical Field

[0001] This invention relates to the field of database query technology, and in particular to a data storage method, electronic device, and storage medium for entity data. Background Technology

[0002] As enterprises deepen their digital transformation, the volume and form of entity data are growing exponentially. Faced with ever-increasing business, entity data databases need to set up multiple servers for querying, retrieving, and storing entity data. However, there are no fixed rules for storing entity data in existing databases. Therefore, when highly matched entity data is stored on different servers, the cost of retrieval increases and the time required also increases. Thus, there is an urgent need for an entity data storage method that can improve retrieval efficiency. Summary of the Invention

[0003] To address the aforementioned technical problems, the technical solution adopted by this invention is as follows:

[0004] A method for storing entity data includes the following steps:

[0005] S100. Obtain the entity data type set A = (A1, A2, ..., A...). i A m ), where the i-th and entity data type list A i =(QA) i A i1 A i2 A ij A in(i) A ij For the j-th entity data information in the i-th entity data type list, the entity data information includes at least: the entity data and the corresponding relationship between the entity data, QA i For A i The corresponding entity type labels are i = 1, 2, ..., m, where m is the number of entity types in the database, and j = 1, 2, ..., n(i), where n(i) is the number of entity data in the i-th entity data type list.

[0006] S200. Obtain the entity type grouping result set U = (U1, U2, ..., Ui, ..., Um) corresponding to A, where the i-th entity type grouping result list Ui = (Ui1, Ui2, ..., Ui...). r , ..., Ui R(i) ), Ui r For A i The corresponding r-th grouping result, r = 1, 2, ..., R(i), where R(i) is A i The corresponding number of grouping results, Ui is the number of groups A according to the preset grouping rules.i each group result in Ui includes at least one entity data group, and each group result in Ui includes A i each entity data information in Ui, and any two entity data groups in the same group result do not include the same entity data information.

[0007] S300, obtain a first priority list set YM=(YM1, YM2, …, YM i , …, YM m ) corresponding to Ui, YM i =(YM i1 , YM i2 , …, YM ir , …, YM iR(i) ) corresponding to Ui, YM ir is the first priority corresponding to Ui r , and M ir satisfies the following conditions:

[0008]

[0009] L i is the number of association relationships corresponding to the entity data in A i , GD ije is the association relationship identifier between A ij and A ie , the association relationship identifier is used to indicate whether there is an association relationship between A ij and A ie , e=1, 2, …, n(i) and A ij ≠A ie , k ij is the number of association relationships corresponding to A ij , k ie is the number of association relationships corresponding to A ie , δ ie ij is the grouping relationship identifier between A ij and A ie , the grouping relationship identifier is used to indicate whether A ij and A ie are in the same entity data group.

[0010] S400, obtain a second priority list E=(E1, E2, …, E i , …, E m ) corresponding to A i , E i is the second priority corresponding to A i , and E i= max(YM i ).

[0011] S500, obtaining a grouping result corresponding to each second priority, to obtain a first entity data group set B=(B1, B2, …, B i , …, B m ), wherein E i corresponds to a first entity data group list B i =(B i1 , B i2 , …, B ix , …, B ip(i) ), B ix is the xth first entity data group in B i , x=1, 2, …, p(i), and p(i) is the number of first entity data groups in B i .

[0012] S600, obtaining a first entity data group identifier list BS=(BS1, BS2, …, BS a , …, BS K ) corresponding to each first entity data group in B, BS a is the ath first entity data group identifier, a=1, 2, …, K, K is the number of first entity data groups, the first entity data group identifier is used for at least indicating the entity data type corresponding to the first entity data group, the first entity data identifiers in the first entity data identifier list are all different, and the matching degree between BS t and BS t+1 is greater than the matching degree between BS t and BS t+2 , t=1, 2, …, K-2.

[0013] S700, when the number F of preset storage domains in the database is greater than or equal to K, obtaining a first storage domain identifier list C=(C1, C2, …, C a , …, C K ), C a is the storage domain identifier of the ath storage domain in the database.

[0014] S800, storing the first entity data in the first entity data group corresponding to BS a and BS a into the storage domain corresponding to C a .

[0015] The present application has at least the following beneficial effects:

[0016] According to the above, by classifying entity data, grouping each classified entity data list, and obtaining the optimal grouping result corresponding to each entity data list, the first entity data group set is obtained according to the optimal grouping result corresponding to each entity data list, the feature corresponding to the first entity data group is extracted to form the first entity data group identifier list corresponding to the first entity data group set, the first entity data identifier and the first entity data group corresponding to the first entity data identifier are stored in the storage domain in the order of the first entity data group identifier list, the same entity data in the same entity data category is in the same first entity data group by obtaining the first entity data group set according to the optimal grouping result, when the entity data needs to be extracted, the queried entity data has a mutual matching relationship, when the storage domain with high matching degree is stored in two different servers, cross-server extraction of the entity data in the storage domain may occur, the first entity data group with high matching degree is stored in the adjacent storage domain, the association degree between the adjacent first entity data groups is the highest, when there are multiple servers, the first entity data group with high matching degree is stored in the same server, when the entity data stored in the storage domain is extracted, frequent cross-server operation is avoided, therefore, it is convenient to extract and query the entity data subsequently, and the time efficiency is improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0018] Figure 1 A flowchart of a data storage method for entity data provided by the embodiments of the present application. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0020] The present application provides a data storage method for entity data, as shown in Figure 1 The method comprises the following steps:

[0021] S100, obtaining an entity data type set A=(A1, A2, …, A i , …, Am ), wherein the i-th entity data type list A i = (QA i , A i1 , A i2 , …, A ij , …, A in(i) ), A ij is the j-th entity data information in the i-th entity data type list, and the entity data information at least includes entity data and an association corresponding to the entity data, QA i is an entity type label corresponding to A i , i = 1, 2, …, m, m is the number of entity types in the database, j = 1, 2, …, n(i), and n(i) is the number of entity data in the i-th entity data type list.

[0022] Specifically, the entity data types are generated according to entities contained in the database, for example, the entity data types can include human, fruits and vegetables, etc.

[0023] Further, the entity type labels can be automatically generated according to the number of entity types, and a person skilled in the art can set the entity type label sequence by himself / herself, for example, using numbers or letters as entity type labels, specifically, the entity type label corresponding to A1 is “1” or “a”, and the entity type label corresponding to A2 is “2” or “b”.

[0024] S200, obtaining an entity type grouping result set U = (U1, U2, …, Ui, …, Um) corresponding to A, wherein the i-th entity type grouping result list Ui = (Ui1, Ui2, …, Ui r , …, Ui R(i) ), Ui r is the r-th grouping result corresponding to A i , r = 1, 2, …, R(i), R(i) is the number of grouping results corresponding to A i , Ui is obtained by grouping the entity data in A i by using a preset grouping rule, each grouping result in Ui includes at least one entity data group, and each grouping result in Ui includes each entity data information in A i , and any two entity data groups in the same grouping result do not include the same entity data information.

[0025] Specifically, the number of entity type grouping results is obtained by using a Fast-Greedy Modularity-Maximization algorithm.

[0026] S300. Obtain the first priority list set YM = (YM1, YM2, ..., YM) corresponding to Ui. i , ..., YM m ), the first priority list YM corresponding to Ui i =(YM i1 YM i2 , ..., YM ir , ..., YM iR(i) YM ir For UI r The corresponding first priority, where M ir Meets the following conditions:

[0027]

[0028] Among them, L i For A i The number of relationships corresponding to entity data in GD ije For A ij With A ie The association identifier is used to represent A. ij With A ie Does there exist a correlation between them, e = 1, 2, ..., n(i) and A ij ≠A ie k ij For A ij The number of corresponding relationships, k ie For A ie The number of corresponding associations, δ ie ij For A ij With A ie The grouping relationship identifier is used to represent A. ij With A ie Whether they are in the same entity data group.

[0029] Specifically, δ ie ij By querying the r-th group result Ui in Ui r China A ij With A ie Do they belong to the same entity data group?

[0030] Furthermore, when A ij With A ie In the same entity data group, δ ix ij =1; when A ij With A ie Located in different entity data groups, δ ix ij =0.

[0031] Further, when there is an association relationship between A ij and A ie , GD ije =1; when there is no association relationship between A ij and A ie , GD ije =0.

[0032] S400, a second priority list E=(E1, E2, …, E i , …, E m ) corresponding to A is obtained, E i is a second priority corresponding to A i , and E i meets the following condition: E i =max(YM i ).

[0033] Specifically, it can be known from the calculation formula of YM ir that the greater the value of the first priority is, the stronger the association relationship between the entity data in each group in the corresponding grouping result of the entity data is, and the stronger the association relationship between the entity data in the entity data group is, the better the grouping result is, so the greater the value of the first priority is, the better the corresponding entity data grouping result is, and therefore, the first priority with the highest priority in the first priority list is selected as the second priority, so that the association relationship between the entity data in the first entity data group is the strongest.

[0034] S500, a grouping result corresponding to each second priority is obtained, and a first entity data group set B=(B1, B2, …, B i , …, B m ) is obtained, wherein E i corresponds to a first entity data group list B i =(B i1 , B i2 , …, B ix , …, B ip(i) ), B ix is the xth first entity data group in B i , x=1, 2, …, p(i), and p(i) is the number of first entity data groups in B i .

[0035] S600, a first entity data group identifier list BS=(BS1, BS2, …, BS a , …, BS K ) corresponding to each first entity data group in B is obtained, BS aa-th first entity data group identifier, a = 1, 2, …, K, K is the number of the first entity data groups, the first entity data group identifier is used at least for representing the entity data type corresponding to the first entity data group, the first entity data identifiers in the first entity data group identifier list are all different, and the BS t has a matching degree greater than the matching degree between the BS t+1 and the BS t , t = 1, 2, …, K-2. t+2

[0036] Specifically, K meets the following condition: K = P(1) + P(2) + …P(i) + … + P(m).

[0037] Further, the first entity data group identifier can also be used for identifying the attribute corresponding to the first entity data group; the embodiment provides an example of the first entity data group identifier corresponding to the first entity data group, for example, the first entity data group identifier corresponding to the first entity data type is (1-1, the attribute corresponding to the first entity data group corresponding to the first entity data type).

[0038] Further, any method for calculating the matching degree between the first entity data group identifiers falls within the protection scope of the present application, which will not be repeated here.

[0039] S700, when the number F of the preset storage domains in the database is greater than or equal to K, a first storage domain identifier list C = (C1, C2, …, C a , …, C K ) is obtained, C a is the storage domain identifier of the a-th storage domain in the database.

[0040] Specifically, the user can set the first storage domain identifier by himself / herself, or the first storage domain identifier can be automatically generated by the processor.

[0041] Further, the first storage domain identifiers in the first storage domain identifier list are all different from the first entity data group identifiers in the first entity data group identifier list.

[0042] S800, the first entity data in the first entity data group corresponding to the BS a and the BS a is stored into the storage domain corresponding to C a .

[0043] ​Specifically, the storage domain corresponds to several servers in a distributed storage manner, and the number of servers is not less than two. Among them, the servers include at least one master server and at least one backup server. The master server is mainly used to search and call the storage domain, and the backup server is used to store the entity data information corresponding to the storage domain.

[0044] The above describes a process where entity data is categorized, and each category of entity data is grouped to obtain the optimal grouping result for each category. Based on this optimal grouping result, a first entity data set is obtained. Features corresponding to the first entity data set are extracted to construct a first entity data set identifier list. The first entity data identifiers and their corresponding first entity data sets are stored in the storage domain in the order listed in the first entity data set identifier list. By obtaining the first entity data set based on the optimal grouping result, the number of similar entities within the same entity data category is reduced. When entity data is stored in the same primary entity data group, and the entity data being retrieved has a matching relationship, if the storage domains with high matching degrees are stored on two different servers, cross-server retrieval of entity data within the storage domain may occur. By storing the primary entity data groups with high matching degrees in adjacent storage domains, the correlation between adjacent primary entity data groups is maximized. When multiple servers exist, storing the primary entity data groups with high matching degrees on the same server avoids frequent cross-server operations when retrieving entity data stored in the storage domain. Therefore, it facilitates subsequent retrieval and querying of entity data, improving time efficiency.

[0045] The entity data stored in the storage domain can be retrieved through the following steps:

[0046] S1. Obtain query conditions, which include at least the entity data type.

[0047] Specifically, the query conditions may also include attributes corresponding to the entity data.

[0048] S2. Obtain a list of specified similarities TD = (TD1, TD2, ..., TD3) between the query conditions and the identifiers of each first entity data group in the first storage domain. a , ..., TD K ), TD a For query conditions and BS a The similarity between them.

[0049] Specifically, those skilled in the art will know that any method for calculating the specified similarity between the query conditions and the identifier of each first entity data group in the first storage domain falls within the protection scope of this invention, and will not be elaborated further here.

[0050] S3, obtaining the specified similarity greater than the preset similarity threshold in the specified similarity list, to obtain the key similarity list ZD=(ZD1, ZD2, …, ZD β , …, ZD γ ), ZD β is the βth key similarity, β=1, 2, …, γ, γ is the number of key similarities.

[0051] Specifically, the similarity threshold can be set by the person skilled in the art according to actual needs.

[0052] Further, the range of the similarity threshold is 60% to 100%.

[0053] S4, obtaining the third storage domain identifier list CZ=(CZ1, CZ2, …, CZ β , …, CZ γ ) corresponding to each final similarity in the key similarity list. β CZ β is the βth third storage domain identifier, and the third storage domain identifier is the first storage domain identifier corresponding to ZD f .

[0054] S5, extracting the storage information in the storage domain corresponding to each third storage domain identifier.

[0055] S900, when F F , obtaining the second entity data group list P={P1, P2, …, P f} after data group fusion according to the preset fusion rule, P a is the fth second entity data group, f=1, 2, …, F.

[0056] Specifically, the second entity data group is obtained by the following steps:

[0057] S910, obtaining the target number K1, K1 meets the following condition: K1=K-F.

[0058] S920, obtaining the number of entity data information in each first entity data group and performing sorting processing, to obtain N=(N1, N2, …, N K , …, N a ), N a is the number of entity data information in the ath first entity data group; N1 K .

[0059] S930, obtaining the third entity data group list D=(D1, D2, …, D a , …, DK ), D a is N a corresponding third entity data group.

[0060] S940, obtain the first K1 third entity data groups in D to obtain a fourth entity data group list M=(M1, M2, …, M c , …, M K1 ), M c is the cth fourth entity data group, c=1, 2, …, K1.

[0061] S950, obtain the third entity data groups in D except the fourth entity data groups to obtain a fifth entity data group list H=(H1, H2, …, H f , …, H F ), H f is the fth fifth entity data group in H.

[0062] Specifically, the 1st fifth entity data group is the third entity data group corresponding to the number of entity data information in the K1+1th first entity data group, the 2nd fifth entity data group is the third entity data group corresponding to the number of entity data information in the K1+2th first entity data group, and so on, and the fth fifth entity data group is the third entity data group corresponding to the number of entity data information in the K1+fth first entity data group.

[0063] S960, obtain the target similarity set DM=(DM1, DM2, …, DM c , …, DM K1 ) corresponding to the fourth entity data group, wherein M c corresponding target similarity list DM c =(DM c1 , DM c2 , …, DM cf , …, DM cF ), DM cf is the target similarity between M c and H f .

[0064] Specifically, DM cf is obtained by the following steps:

[0065] S961, obtain the feature vector list set LM={LM1, LM2, …, LM c , …, LM K1} corresponding to the fourth entity data group in M, wherein M c corresponding feature vector list LM c =(LM c1 , LM c2 , …, LMcw ,..., LM cz ), LM cw is the feature vector corresponding to the fourth entity data group. c is the wth feature vector extracted from M, w = 1, 2,..., z, and z is the number of feature types corresponding to the fourth entity data group.

[0066] Specifically, the feature types of the entity data group at least include: the type corresponding to the entity data group and the attribute corresponding to the entity data group.

[0067] Further, those skilled in the art know that any method for extracting the feature types of the entity data group falls within the protection scope of the present application, and will not be repeated here.

[0068] Further, the feature vector corresponding to the fourth entity data group is obtained according to the feature types corresponding to the fourth entity data group. Those skilled in the art know that any method for obtaining the feature vector corresponding to the fourth entity data group according to the feature types corresponding to the fourth entity data group falls within the protection scope of the present application, and will not be repeated here.

[0069] S963, obtaining the feature vector list set LH = {LH1, LH2,..., LH f ,..., LH F} corresponding to the fifth entity data group in H, wherein H f is the number of feature types corresponding to the fifth entity data group. f LH f1 = {LH f2 , LH fw ,..., LH fz}, LH fw is the feature vector corresponding to H f .

[0070] S965, obtaining DM c based on LM f and LH cf , wherein DM cf meets the following condition:

[0071]

[0072] S1000, obtaining the second entity data group identifier list PS = (PS1, PS2,..., PS f ,..., PS F ) corresponding to each second entity group in P, wherein PS f is the entity data identifier corresponding to the fth second entity data group, the second entity data group is used at least to represent the entity data type corresponding to the second entity data group, the second entity data identifiers in the second entity data group identifier list are all different, and PS q is different from PSq+1 between the PSs is greater than PS q between the PSs q+2 between the PSs, q=1, 2, …, F-2.

[0073] S1100, a second storage domain identifier list CP={CP1, CP2, …, CP f , …, CP F} is acquired, CP f is a storage domain identifier of the fth storage domain in the database.

[0074] S1200, the first entity data PS f and the second entity data in the second entity data group corresponding to the PS f are stored into the storage domain corresponding to CP f .

[0075] Specifically, the method of extracting the storage domain corresponding to the second entity data group is the same as the method of extracting the storage domain corresponding to the first entity data group.

[0076] For the method of acquiring the second entity data group, the technical effects are the same as those of the first entity data group, and thus will not be described herein.

[0077] Embodiments of the present application further provide a non-transitory computer readable storage medium, which can be arranged in an electronic device to save at least one instruction or at least one program related to a method in the method embodiments, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the method provided by the above embodiments.

[0078] Embodiments of the present application further provide an electronic device, which comprises a processor and the aforementioned non-transitory computer readable storage medium.

[0079] Although some specific embodiments of the present application have been described in detail by examples, those skilled in the art should understand that the above examples are only for illustration, but not for limiting the scope of the present application. Those skilled in the art should also understand that various modifications can be made to the embodiments without departing from the scope and spirit of the present application. The scope of the present application is defined by the appended claims.

Claims

1. A data storage method for entity data, characterized by, The method comprises the following steps: S100. Obtain the entity data type set A = (A1, A2, ..., A...). i A m ), where the i-th entity data type list A i =(QA i A i1 A i2 A ij A in（i） A ij For the j-th entity data information in the i-th entity data type list, the entity data information includes at least: entity data and the corresponding association relationship of the entity data, QA i For A i The corresponding entity type label, i=1,2,…,m, where m is the number of entity types in the database, j=1,2,…,n(i), where n(i) is the number of entity data in the i-th entity data type list; S200, acquire an entity type grouping result set U=(U1, U2,..., Ui,..., Um) corresponding to A, wherein the i-th entity type grouping result list Ui=(Ui1, Ui2,..., Ui r ,..., Ui R（i） ) of A, r=1, 2,..., R(i), R(i) is the number of grouping results corresponding to A, Ui is obtained by grouping the entity data in A through a preset grouping rule, each grouping result in Ui includes at least one entity data group, and each grouping result in Ui includes each entity data information in A, and any two entity data groups in the same grouping result do not include the same entity data information. r i i i i ​​​​​ S300, Obtain the first priority list set YM = (YM1, YM2, ..., YM) corresponding to Ui. i , ..., YM m ), the first priority list YM corresponding to Ui i =(YM i1 YM i2 , ..., YM ir , ..., YM iR（i） ), YM ir For UI r The corresponding first priority, where M ir Meets the following conditions: ； Among them, L i For A i The number of relationships corresponding to entity data in GD ije For A ij With A ie The association identifier is used to represent A. ij With A ie Does there exist a correlation between them, e=1,2,…,n(i) and A? ij ≠A ie k ij For A ij The number of corresponding relationships, k ie For A ie The number of corresponding associations, δ ie ij For A ij With A ie The grouping relationship identifier is used to represent A. ij With A ie Are they in the same entity data group? S400, acquiring a second priority list E=(E1, E2, …, En) corresponding to A, where E is a second priority corresponding to A, and E satisfies the following conditions: E1> E2> …> En, and E1= max(YM1), E2= max(YM2), …, En= max(YMn). i m i i i i i ​​​​​​​ S500. Obtain the grouping result corresponding to each second priority, and obtain the first entity data set B = (B1, B2, ..., B...). i B m ), where E i The corresponding first entity data group list B i = (B i1 B i2 B ix B ip（i） ), B ix For B i In the x-th first entity data group, x=1,2,…,p(i), p(i) is B i The number of the first entity data group in the middle; S600, obtaining a first entity data group identifier list BS=(BS1, BS2, …, BS a , …, BS K ) corresponding to each first entity data group in B, BS a is an athfirst entity data group identifier, a=1, 2, …, K, K is the number of first entity data groups, the first entity data group identifier is used at least for representing the entity data type corresponding to the first entity data group, the first entity data identifiers in the first entity data group identifier list are all different, and the matching degree between BS t and BS t+1 is greater than the matching degree between BS t and BS t+2 , t=1, 2, …, K-2; S700. When the number of preset storage domains in the database F ≥ K, obtain the first storage domain identifier list C = (C1, C2, ..., C...). a C K ), C a This is the storage domain identifier for the a-th storage domain in the database; S800, the BS a and the BS a the first entity data in the corresponding first entity data set is stored to the C a in the corresponding storage domain; S900、When F f , the first entity data set in B that meets the preset condition is fused according to a preset fusion rule to obtain a list of fused second entity data sets P = {P1, P2, …, P F} where P f is the fth second entity data set, f = 1, 2, …, F. S1000、When F f , the first entity data set in B that meets the preset condition is fused according to a preset fusion rule to obtain a list of fused second entity data sets P = {P1, P2, …, P F} where P f is the fth second entity data set, f = 1, 2, …, F. S1000、acquire a second entity data group identification list PS= (PS1, PS2, …, PSF) corresponding to each second entity group in P, wherein PSf is an entity data identification corresponding to the fth second entity data group, the second entity data group is used for representing at least an entity data type corresponding to the second entity data group, the second entity data identifications in the second entity data group identification list are all different, and the matching degree between PSf and PSf-1 is greater than the matching degree between PSf and PSf+1, q = 1, 2, …, F-2. f F f q q+1 q q+2 ​​​​​​​ S1100, acquire a second storage domain identification list CP={CP1, CP2, …, CP f , …, CP F}, CP f is the storage domain identification of the fth storage domain in the database; S1200, PS f And PS f The second entity data in the corresponding second entity data group is stored in CP. f Within the corresponding storage domain.

2. The method of claim 1, wherein, The second entity data set is obtained by the following steps: S910, obtaining a target quantity K1, K1 meets the following condition: K1=K-F; S920. Obtain the number of entity data information in each first entity data group and sort them to obtain N = (N1, N2, ..., N...). a ,…,N K ), N a Let N be the number of entity data information within the a-th first entity data group; N1 < N2 < ... < N a <...<N K ; S930, acquire a third entity data group list D=(D1, D2, …, D a , K D a is the N a corresponding third entity data group; S940, obtaining K1 third entity data groups in D, to obtain a fourth entity data group list M=(M1, M2, …, M c , K1 M c c is the cth fourth entity data group, c=1, 2, …, K1; S950, obtaining the third entity data set in D except the fourth entity data set, to obtain a fifth entity data set list H=(H1, H2, …, H f , F H f is the fth fifth entity data set in H; S960. Obtain the target similarity set DM = (DM1, DM2, ..., DM) corresponding to the fourth entity data group. c , ..., DM K1 ), where M c Corresponding target similarity list DM c =(DM c1 DM c2 , ..., DM cf , ..., DM cF ), DM cf For M c With H f Target similarity between them; S970. Obtain the key similarity list GM = (GM1, GM2, ..., GM...). c , ..., GM K1 ), GM c For M c The corresponding key similarity, and GM c Meets the following conditions: GM c =max(DM) cf ); S980、M c with GM c The fifth entity data set is fused to obtain a fused second entity data set list P.

3. The method of claim 2, wherein, DM cf By the following steps: S961、acquire the feature vector list set LM={LM1, LM2, …, LM corresponding to the fourth entity data group in M c , …, LM K1}, wherein M c corresponds to the feature vector list LM c =(LM c1 , LM c2 , …, LM cw , …, LM cz ), LM cw is the wth feature vector extracted from M c , w=1, 2, …, z, and z is the number of feature types corresponding to the fourth entity data group. S963、acquire the feature vector list set LH={LH1, LH2, …, LH f} corresponding to the fifth entity data group in H, wherein H F corresponds to the feature vector list LH f ={LH f , LH f1 , …, LH f2 , …, LH fw , …, LH fz}, LH fw is the wth feature vector extracted for H f ; S965, based on LM c with LH f , obtaining DM cf where DM cf satisfies the following conditions:

4. The method of claim 3, wherein, The feature type of the entity data set at least includes: a type corresponding to the entity data set and an attribute corresponding to the entity data set.

5. The method of claim 1, wherein, The entity data stored in the storage domain is called by the following steps: S1, obtaining a query condition, the query condition at least includes: an entity data type; S2, obtaining a specified similarity list TD= (TD1, TD2, …, TDn) between the query condition and each first entity data group identifier in the first storage domain, TD being the similarity between the query condition and BS a , …, TD K n), TD a being the similarity between the query condition and BS a ​ S3, obtaining the specified similarity greater than the preset similarity threshold in the specified similarity list, to obtain the key similarity list ZD= (ZD1, ZD2, …, ZD β , …, ZD γ ), ZD β is the βth key similarity, β=1, 2, …, γ, and γ is the number of key similarities; S4, obtaining, in the key similarity list, a third storage domain identifier list CZ=(CZ1, CZ2, …, CZn) corresponding to each final similarity, wherein CZ is a βth third storage domain identifier, and the third storage domain identifier is a first storage domain identifier corresponding to ZD β . γ . β . β . S5, extracting the storage information in the storage domain corresponding to each third storage domain identifier.

6. The method of claim 1, wherein, When A ij is in the same entity data set, δ ie = 1. ix ij = 1. When A ij with A ie are in different entity data sets, δ ix ij = 0.

7. The method of claim 1, wherein, K meets the following condition: K=P(1)+P(2)+…P(i)+…+P(m).

8. A non-transitory computer readable storage medium, the storage medium storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by a processor to implement the method of any one of claims 1-7.

9. An electronic device, comprising: A processor and the non-transitory computer readable storage medium of claim 8.

Images