Method and device for knowledge evolution feature mining, electronic equipment and storage medium

Abstract

The application provides a knowledge evolution feature mining method and device, electronic equipment and a storage medium, comprising: acquiring a first relationship between a plurality of learners and a plurality of knowledge entities, a second relationship between a plurality of knowledge entities and a plurality of knowledge entities, and a third relationship between a plurality of learners and a plurality of learners at a t time under a target knowledge graph; based on the first relationship, the second relationship, the third relationship and the target knowledge graph, an evolution key path obtained in an evolution process from the t time to the t+1 time is determined; according to the evolution key path, a target subject corresponding to each evolution key path is determined; a change trend value of all evolution key paths associated with the target subject in the target knowledge graph is acquired, wherein the change trend value is used to represent whether the evolution characteristic corresponding to the evolution key path is established; using the change trend value and the second-order relationship number of the target subject at the t time, an evolution feature value corresponding to the evolution characteristic is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers, and in particular to a knowledge evolution feature mining method and device, an electronic device, and a storage medium. BACKGROUND

[0002] With the development of Internet technology, a large amount of text data has emerged in the network. How to mine rules from text data has become a research hotspot. Among numerous data mining technologies, knowledge graphs have developed rapidly in recent years and have become a main auxiliary tool for building intelligent question answering systems and recommendation systems.

[0003] Online learning communities contain rich knowledge shared and created by individuals and groups. It is very important to mine this knowledge and build a knowledge graph that can represent the knowledge of the community. Not only can it effectively support personalized learning recommendations and intelligent search for community members, but also can improve the learning experience and efficiency of online communities, and promote the creation and development of community knowledge. However, the knowledge in the community is often scattered in the personal and interactive spaces of the community in an unstructured and fragmented way, and constantly evolves with the sharing, interaction and creation of community members, which makes the construction of knowledge graphs in online learning communities challenging.

[0004] Although some technologies have explored knowledge mining methods for such online learning communities, there is a lack of a community knowledge graph construction method that comprehensively considers the development and production characteristics of community knowledge, especially ignoring the feature that the knowledge in online learning communities evolves with the sharing, interaction and creation of community members.

[0005] Therefore, the related art lacks mining of the evolution of community knowledge in constructing a community knowledge graph, resulting in inaccurate prediction results for various application scenarios in an online learning community based on a knowledge graph. SUMMARY

[0006] The present application provides a knowledge evolution feature mining method and device, an electronic device and a storage medium to at least solve the problem that the related art lacks mining of the evolution of community knowledge in constructing a community knowledge graph, resulting in inaccurate prediction results for various application scenarios in an online learning community based on a knowledge graph.

[0007] According to an aspect of an embodiment of the present application, a knowledge evolution feature mining method is provided, which includes:

[0008] Obtaining a first relationship between a plurality of learners and a plurality of knowledge entities, a second relationship between a plurality of knowledge entities and a plurality of knowledge entities, and a third relationship between a plurality of learners and a plurality of learners at a t-th moment under a target knowledge graph;

[0009] Based on the first relationship, the second relationship, the third relationship, and the target knowledge graph, the evolutionary critical path obtained from the evolution process from time t to time t+1 is determined, wherein the evolutionary critical path is a critical path under multiple preset evolutionary modes;

[0010] Based on the aforementioned evolutionary critical paths, determine the target subject corresponding to each of the aforementioned evolutionary critical paths;

[0011] Obtain the change trend values ​​of all the evolutionary key paths associated with the target subject in the target knowledge graph, wherein the change trend value represents the amount of change of the evolutionary key path, and is used to characterize whether the evolutionary characteristics corresponding to the evolutionary key path are valid;

[0012] Using the trend value and the second-order relation coefficient of the target subject at time t, the evolutionary characteristic value corresponding to the evolutionary characteristics is obtained.

[0013] According to another aspect of the embodiments of this application, an apparatus for knowledge evolution feature mining is also provided, the apparatus comprising:

[0014] The first acquisition module is used to acquire the first relationship between multiple learners and multiple knowledge entities, the second relationship between multiple knowledge entities, and the third relationship between multiple learners at time t under the target knowledge graph.

[0015] The first determining module is used to determine the evolutionary critical path obtained from the evolution process from time t to time t+1 based on the first relationship, the second relationship, the third relationship and the target knowledge graph, wherein the evolutionary critical path is a critical path under multiple preset evolution modes;

[0016] The second determining module is used to determine the target subject corresponding to each evolutionary critical path based on the evolutionary critical path;

[0017] The second acquisition module is used to acquire the change trend values ​​of all the evolutionary key paths associated with the target subject in the target knowledge graph, wherein the change trend value represents the change amount of the evolutionary key path, and is used to characterize whether the evolutionary characteristics corresponding to the evolutionary key path are valid.

[0018] The first obtaining module is used to obtain the evolutionary characteristic value corresponding to the evolutionary characteristics by using the change trend value and the second-order relation coefficient of the target subject at time t.

[0019] According to another aspect of the embodiments of this application, an electronic device is also provided, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; wherein the memory is used to store a computer program; and the processor is used to execute the method steps of any of the above embodiments by running the computer program stored in the memory.

[0020] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored therein, wherein the computer program is configured to execute the method steps of any of the above embodiments when running.

[0021] In this embodiment, based on the relationships between learners, between learners and knowledge entities, and between knowledge entities at time t in the knowledge graph, the evolutionary critical path at time t+1 is obtained. The corresponding target subject is determined based on the evolutionary critical path, and then all trend values ​​associated with each target subject are obtained. Based on these trend values ​​and the second-order relational coefficient of the target subject at time t, the evolutionary characteristic value of the evolutionary feature corresponding to the evolutionary critical path is determined. After obtaining the evolutionary characteristic value based on the above, it can support various prediction application scenarios in the subsequent knowledge graph-based online learning community, obtaining relatively accurate prediction results. This solves the problem that related technologies lack the mining of the community knowledge evolution feature in constructing community knowledge graphs, leading to inaccurate prediction results for various application scenarios in the subsequent knowledge graph-based online learning community. Attached Figure Description

[0022] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the hardware environment for an optional knowledge evolution feature mining method according to an embodiment of this application;

[0025] Figure 2 This is a flowchart illustrating an optional knowledge evolution feature mining method according to an embodiment of this application;

[0026] Figure 3This is a schematic diagram of an optional learner-mediated transitivity evolutionary critical path evolutionary mode according to an embodiment of this application;

[0027] Figure 4 This is a schematic diagram of an optional knowledge-mediated transitivity evolutionary critical path evolutionary mode according to an embodiment of this application;

[0028] Figure 5 This is a schematic diagram of an optional learner reciprocity evolution critical path evolution mode according to an embodiment of this application;

[0029] Figure 6 This is a schematic diagram of an optional basic relationship structure between knowledge entities according to an embodiment of this application;

[0030] Figure 7 This is a structural block diagram of an optional knowledge evolution feature mining apparatus according to an embodiment of this application;

[0031] Figure 8 This is a structural block diagram of an optional electronic device according to an embodiment of this application. Detailed Implementation

[0032] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0033] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar learners and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0034] According to one aspect of the embodiments of this application, a method for knowledge evolution feature mining is provided. Optionally, in this embodiment, the above-described knowledge evolution feature mining method can be applied to, for example... Figure 1 In the hardware environment shown. For example... Figure 1As shown, terminal 102 may include memory 104, processor 106, and display 108 (optional component). Terminal 102 can communicate with server 112 via network 110. Server 112 can provide services (such as application services) to the terminal or clients installed on the terminal. Database 114 can be set up on or independently of server 112 to provide data storage services to server 112. In addition, server 112 may run a processing engine 116, which can be used to execute the steps performed by server 112.

[0035] Optionally, terminal 102 may be, but is not limited to, a terminal capable of computing data, such as a mobile terminal (e.g., mobile phone, tablet computer), laptop computer, PC (Personal Computer), etc. The aforementioned network may include, but is not limited to, a wireless network or a wired network. The wireless network includes Bluetooth, Wi-Fi (Wireless Fidelity), and other networks that enable wireless communication. The aforementioned wired network may include, but is not limited to, a wide area network (WAN), a metropolitan area network (MAN), and a local area network (LAN). The aforementioned server 112 may include, but is not limited to, any hardware device capable of computing.

[0036] Furthermore, in this embodiment, the above-described knowledge evolution feature mining method can also be applied to, but is not limited to, a powerful independent processing device without data interaction. For example, the processing device can be, but is not limited to, a powerful terminal device; that is, the various operations in the above-described knowledge evolution feature mining method can be integrated into a single independent processing device. The above is merely an example, and no limitation is made in this embodiment.

[0037] Optionally, in this embodiment, the above-described knowledge evolution feature mining method can be executed by server 112, by terminal 102, or by both server 112 and terminal 102. The knowledge evolution feature mining method of this embodiment can also be executed by a client installed on terminal 102.

[0038] Taking the server side as an example, Figure 2 This is a flowchart illustrating an optional knowledge evolution feature mining method according to an embodiment of this application, as shown below. Figure 2 As shown, the process of this method may include the following steps:

[0039] Step S201: Obtain the first relationship between multiple learners and multiple knowledge entities, the second relationship between multiple knowledge entities, and the third relationship between multiple learners at time t under the target knowledge graph.

[0040] Step S202: Based on the first relation, the second relation, the third relation and the target knowledge graph, determine the evolutionary critical path obtained from the evolution process from time t to time t+1, wherein the evolutionary critical path is the critical path under multiple preset evolution modes;

[0041] Step S203: Based on the evolutionary critical path, determine the target subject corresponding to each evolutionary critical path;

[0042] Step S204: Obtain the change trend values ​​of all evolutionary critical paths associated with the target subject in the target knowledge graph. The change trend value represents the amount of change of the evolutionary critical path, which is used to characterize whether the evolutionary characteristics corresponding to the evolutionary critical path are valid.

[0043] Step S205: Using the trend value and the second-order relation coefficient of the target subject at time t, obtain the evolutionary characteristic value corresponding to the evolutionary characteristics.

[0044] Optionally, an online learning community knowledge graph (also known as a target knowledge graph) is a visual representation of community knowledge that shows the development process and structural relationships of community knowledge, formed as learners share, interact, and create within the community. This graph includes two types of nodes: learners and knowledge entities; and three networks: the learner's social network, the knowledge entity structure network, and the relationship network between learners and knowledge entities. Learners are the main participants in the online learning community, and knowledge entities are keywords existing in the learning resource texts shared and created by learners.

[0045] To obtain evolutionary features, this application requires the evolutionary feature values ​​generated during the evolution of the target knowledge graph from time t to time t+1. Specifically, it first obtains the first relationship between multiple learners and multiple knowledge entities, the second relationship between multiple knowledge entities, and the third relationship between multiple learners at time t, such as... Figure 3 As shown, w t Learner 1 (actor1) and Learner 2 (actor2) have a social relationship, and Learner 2 has a relationship with knowledge entity 1 (word1); for example Figure 4 As shown, w t Learner 1 is related to knowledge entity 1, and knowledge entity 1 is related to knowledge entity 2 (word2); for example... Figure 5 As shown, w t Learner 1 is connected to knowledge entity 1 under the window, and learner 1 is connected to learner 2.

[0046] Having established the first, second, and third relations, we can deduce, based on the target knowledge graph, the result at time t+1, i.e.Figure 3 In w t+1 Learner 1 exhibits a tendency to strengthen their relationship with knowledge entity 1. That is, during the evolutionary process, learner 1 enhances their relationship with knowledge entity 1 through learner 2. This represents the learner's ability to deepen their understanding of knowledge through other learners. Correspondingly, in Figure 4 Chinese w t+1 Learner 1 exhibits a tendency to strengthen their relationship with knowledge entity 2. That is, during the evolutionary process, learner 1 enhances their relationship with knowledge entity 2 through knowledge entity 1. This represents the learner's ability to deepen their understanding of other knowledge entities through knowledge entities. Correspondingly, in Figure 5 In w t+1 The window indicates that learner 2 has a tendency to strengthen its relationship with learner 1 and knowledge entity 1. That is, interaction between learners strengthens their social relationships and their relationships with related knowledge entities. This represents the ability of learners to enhance their understanding of a particular knowledge entity through interactive learning.

[0047] Therefore, by Figure 3 The relationship line between learner 1 and knowledge entity 1 can be called the evolutionary critical path obtained from time t to time t+1. This evolutionary critical path corresponds to... Figure 3 The key pathway of learner-mediated transitive evolutionary patterns; similarly, by Figure 4 The relationship line between learner 1 and knowledge entity 2 can be called the evolutionary critical path obtained from time t to time t+1. This evolutionary critical path corresponds to... Figure 4 The key path of the knowledge-mediated transmission evolutionary model; similarly, by Figure 5 The relationship line between learner 2, learner 1, and knowledge entity 1 can be called the evolutionary critical path obtained from time t to time t+1. This evolutionary critical path corresponds to... Figure 5 The key path of the learner reciprocity evolution pattern.

[0048] Having identified the aforementioned critical evolutionary paths, the target entity corresponding to each critical evolutionary path is determined based on these paths. For example, Figure 3 The key path of “learner-mediated transitivity” is “actor1->word1”, with actor1 as the target subject; Figure 4 The key path of "knowledge entity mediation and transitivity" is "actor1->word2", and the target subject is actor1; Figure 5 The key paths for “learner reciprocity” are “actor2->actor1” and “actor2->word1”, with actor2 as the target subject.

[0049] After identifying the target entity, since the target knowledge graph is composed of many nodes and node network relationship lines, by taking the target entity as one of the nodes, all the evolutionary critical paths associated with it can be obtained based on the target knowledge graph. Then, the change trend value Δp corresponding to all evolutionary critical paths based on the target entity can be calculated.

[0050] Where, Δp=p t+1 -p t

[0051] t represents time, t+1 represents the next time point with t as the current time, p t p represents the value of the relationship between learners or between learners and knowledge entities, as represented by the critical path of evolution at time t. t+1 This represents the relationship value between learners or between learners and knowledge entities as characterized by the evolutionary critical path at time t+1.

[0052] Furthermore, a positive Δp indicates that the evolutionary characteristic corresponding to a critical evolutionary path holds true. A negative Δp indicates that the evolutionary characteristic corresponding to a critical evolutionary path has deteriorated. A Δp of 0 indicates that the evolutionary characteristic corresponding to a critical evolutionary path remains unchanged.

[0053] Obtain all accumulated Δp values ​​and divide them by the target entity, for example, when the target entity is i, at time t, i.e., w t The second-order relation N under the window it That is, the evolutionary eigenvalues ​​of the corresponding properties:

[0054]

[0055] When the evolutionary trait value is positive, the greater the corresponding evolutionary trend, indicating a more pronounced trend. A negative evolutionary trait value means that the evolutionary characteristic has not strengthened during evolution, but rather weakened. A zero evolutionary trait value means that the evolutionary characteristic has not changed during evolution.

[0056] In this embodiment, based on the relationships between learners, between learners and knowledge entities, and between knowledge entities at time t in the knowledge graph, the evolutionary critical path at time t+1 is obtained. The corresponding target subject is determined based on the evolutionary critical path, and then all trend values ​​associated with each target subject are obtained. Based on these trend values ​​and the second-order relational coefficient of the target subject at time t, the evolutionary characteristic value of the evolutionary feature corresponding to the evolutionary critical path is determined. After obtaining the evolutionary characteristic value based on the above, it can support various prediction application scenarios in the subsequent knowledge graph-based online learning community, obtaining relatively accurate prediction results. This solves the problem that related technologies lack the mining of the community knowledge evolution feature in constructing community knowledge graphs, leading to inaccurate prediction results for various application scenarios in the subsequent knowledge graph-based online learning community.

[0057] As an optional embodiment, the change trend values ​​of all evolutionary critical paths corresponding to the target subject in the target knowledge graph are obtained, including:

[0058] The preset calculation formula for relation extraction is determined based on the critical evolution path;

[0059] The changing trend value of each evolutionary critical path is obtained according to the preset calculation formula;

[0060] Obtain all trend values ​​corresponding to the target subject.

[0061] Optionally, in this embodiment of the application, relation extraction includes the extraction of social network relations between learners and the extraction of relations between learners and knowledge entities, and there are preset calculation formulas for these relation extractions.

[0062] Based on the evolutionary critical path, it can be determined whether the evolutionary critical path represents the social network relationship between learners or the relationship between learners and knowledge entities. Then, the currently applicable preset calculation formula is determined. Then, based on the determined preset calculation formula, the first relationship value at time t and the second relationship value at time t+1 are obtained respectively. The difference between the two is calculated to obtain the trend value.

[0063] The same calculation method is used to calculate the trend values ​​of all evolutionary critical paths related to the target subject, and then all trend values ​​are obtained.

[0064] As an optional embodiment, a preset calculation formula for determining relation extraction based on the evolutionary critical path includes:

[0065] If the target subject at one end of the evolutionary critical path is the learner and the target subject at the other end is the learner, then the calculation formula corresponding to the learner's social network relationship will be extracted and used as the preset calculation formula.

[0066] If the target subject at one end of the evolutionary critical path is the learner and the target subject at the other end is the knowledge entity, then the calculation formula corresponding to the relationship between the learner and the knowledge entity is extracted and used as the preset calculation formula.

[0067] Optionally, such as Figure 3 As shown, the critical path of evolution is "actor1->word1", that is... Figure 3 The dotted arrow in the diagram indicates that the evolutionary critical path connects the target subject actor1 and knowledge entity1. In this case, the corresponding calculation formula for the relationship between the learner and the knowledge entity is extracted and used as the preset calculation formula for calculating the change trend value of the evolutionary critical path.

[0068] like Figure 4 As shown, the critical path of evolution is "actor1->word2", that is... Figure 3 The dotted arrow in the diagram indicates that the evolutionary critical path connects the target subject actor1 and knowledge entity1. In this case, the corresponding calculation formula for the relationship between the learner and the knowledge entity is extracted and used as the preset calculation formula for calculating the change trend value of the evolutionary critical path.

[0069] like Figure 5 As shown, the critical evolutionary paths are "actor2->actor1" and "actor2->word1", that is... Figure 5 The two dashed arrows in the diagram represent the evolutionary critical path connecting the target actor (actor2) and learner actor (actor1), and the target actor (actor2) and knowledge entity (actuary 1). At this point, there are a preset number of calculation formulas, which are the calculation formulas corresponding to the extraction of the relationship between the learner and the knowledge entity, and the calculation formulas corresponding to the extraction of the relationship between the learner and the knowledge entity.

[0070] As an optional embodiment, obtaining the first relation value of the evolutionary critical path at time t, calculated by a preset formula, includes:

[0071] When the preset calculation formula is used to extract the corresponding calculation formula for the learner's social network relationship, the interaction behavior between a preset number of learners associated with the evolutionary key path at time t is obtained to obtain the first interaction feature;

[0072] The first interaction feature is preprocessed to obtain the preprocessed second interaction feature;

[0073] The first relationship value is obtained by using the second interaction feature and the first weight;

[0074] or,

[0075] When the preset calculation formula is the calculation formula for the relationship between learners and knowledge entities, the interaction behavior between learners and knowledge entities associated with the evolutionary key path at time t is obtained to obtain the third interaction feature.

[0076] The third interaction feature is preprocessed to obtain the preprocessed fourth interaction feature;

[0077] The first relation value is obtained by using the fourth interaction feature and the second weight.

[0078] Optionally, when extracting social network relationships among learners, the corresponding preset calculation formula is the calculation formula for extracting social network relationships among learners. At this time, the preset number of evolutionary key path connections at time t is obtained, such as the interaction behaviors between two learners (it can be understood that there must be a target subject included at present) such as likes, comments and follows, to obtain the first interaction feature.

[0079] The first interaction feature is then preprocessed, for example, by performing necessary transformations: logarithmic transformation and normalization, to obtain the preprocessed second interaction feature.

[0080] The weights are calculated using the entropy weight method to obtain the first weight, which is then used as the adjustment value for the second interaction feature. Finally, based on the second interaction feature and the first weight, the first relationship value is obtained, calculated using the following formula:

[0081] p t = w0*f0 + w1*f1 + ... + w n *f n

[0082] Where f0, f1, ..., f n Representing the second interaction feature, w0, w1, ..., w n The weights represent the weights of each second interaction feature.

[0083] or,

[0084] When extracting the relationship between learners and knowledge entities, the corresponding preset calculation formula is the calculation formula for the relationship between learners and knowledge entities. At this time, the interaction behavior between learners and knowledge entities in the evolutionary critical path connection at time t is obtained, such as the number of views, the number of shares, the degree of familiarity and mastery, etc., and the corresponding third interaction feature is obtained.

[0085] The third interaction feature is preprocessed, such as by logarithmic transformation and normalization, to obtain the preprocessed fourth interaction feature.

[0086] The weights are calculated using the entropy weight method to obtain the second weight, which is then used as the adjustment value for the fourth interaction feature. Finally, based on the fourth interaction feature and the second weight, the first relationship value is obtained, calculated using the following formula:

[0087] p t =w 0' *f 0’ +w 1' *f 1’ +...+w n' *f n’

[0088] Among them, f 0' ,f 1' ...f n' Representing the fourth interaction feature, w 0' ,w 1' ...w n' The weights represent the weights of each fourth interaction feature.

[0089] As an optional embodiment, obtaining the second relation value of the evolutionary critical path at time t+1, calculated by a preset formula, includes:

[0090] When the preset calculation formula is used to extract the corresponding calculation formula for the learner's social network relationship, the interaction behavior between a preset number of learners associated with the evolutionary key path is obtained at time t+1, and the fifth interaction feature is obtained.

[0091] The fifth interaction feature is preprocessed to obtain the preprocessed sixth interaction feature;

[0092] The second relation value is obtained by using the sixth interaction feature and the first weight;

[0093] or,

[0094] When the preset calculation formula is the corresponding calculation formula for the relationship between learners and knowledge entities, the interaction behavior between learners and knowledge entities associated with the evolutionary key path at time t+1 is obtained, and the seventh interaction feature is obtained.

[0095] The seventh interaction feature is preprocessed to obtain the preprocessed eighth interaction feature.

[0096] The second relation value is obtained by using the eighth interaction feature and the second weight.

[0097] Optionally, in this embodiment, the second relation value at time t+1 is obtained using the same calculation method as the first relation value at time t. The calculation formulas are as follows:

[0098] When the preset calculation formula is the calculation formula corresponding to the learner's social network relationship extraction:

[0099] p t+1 = w0*f0 + w1*f1 + ... + w n *f n

[0100] Where f0, f1, ..., f n Representing the sixth interaction feature, w0, w1, ..., w n The weights represent the weights of each sixth interaction feature.

[0101] or,

[0102] When the preset calculation formula is the calculation formula corresponding to the relationship between learners and knowledge entities:

[0103] p t+1 =w 0' *f 0’ +w 1' *f 1’ +...+w n' *f n’

[0104] Among them, f 0' ,f 1' ...f n' Representing the eighth interaction feature, w 0' ,w 1' ...w n' The weights represent the weights of each eighth interaction feature.

[0105] The calculation process for the second key value can be found in the calculation process for the first relation value at time t in the above embodiment. Here, time t is simply replaced with time t+1. The process of obtaining interaction features and preprocessing is the same and will not be repeated here.

[0106] As an optional embodiment, before obtaining the first relationship between multiple learners and multiple knowledge entities, the second relationship between multiple knowledge entities, and the third relationship between multiple learners at time t under the target knowledge graph, wherein the method further includes:

[0107] Identify participants in the target scenario as learners;

[0108] Knowledge entities are identified and extracted from the target text, and hierarchical relationships between knowledge entities are constructed based on the target text to obtain the basic relationship structure;

[0109] Based on the average distance and corresponding cosine similarity between any two adjacent knowledge entities in the target text, the association between the two knowledge entities to be optimized is deleted, and the optimized target basic relationship structure is obtained.

[0110] Based on the relationships between learners and knowledge entities and the target-based relationship structure between knowledge entities, a target knowledge graph is obtained.

[0111] Optionally, as described in the preceding embodiments, the target knowledge graph includes two types of nodes: learners and knowledge entities, and three types of networks: the social network among learners, the structural network among knowledge entities, and the relational network between learners and knowledge entities. Therefore, in order to make the target knowledge graph more realistic and reduce redundant relational networks, in this embodiment, the relational network is optimized by pruning the relational edges through the extraction of relationships between knowledge entities.

[0112] Specifically, the learning resources shared and created by participants (i.e. learners) in the target scenario, such as an online learning community, are taken as the target text. Keywords are identified from the target text using deep learning methods, such as BERT+LSTM, to obtain knowledge entities.

[0113] After obtaining multiple knowledge entities, the relationships between these knowledge entities are constructed. For example... Figure 6 As shown, a hierarchical relationship is used to represent this. The rule is as follows: if the set of texts in which knowledge point 1 appears contains the set in which another knowledge point 2 appears, then knowledge point 1 is a superior knowledge point of knowledge point 2.

[0114] Figure 6 In this context, the root node represents all documents {d1, d2, d3, d4, d5} and is a virtual node. For kg1, kg2, and kg3, it is a subordinate node of the root node. For kg4, it appears in d1, and since the knowledge points of the text including d1 are kg1 and kg2, kg4 is a subordinate knowledge point of kg1 and kg2. Similarly, kg5 is a subordinate knowledge point of kg3. This leads to or constructs a set of documents such as... Figure 6 The basic relational structure is shown.

[0115] After obtaining the basic relation structure, some meaningless hierarchical relations will be generated, so optimization is performed through pruning. Specifically, a predetermined number of adjacent entities are obtained, such as the average distance and corresponding cosine similarity between two adjacent knowledge entities in the target text, to obtain...

[0116]

[0117] Where Mean d(kg1,kg2) refers to the average character distance between the two knowledge entities in the target document, Max d(kg1,kg2) refers to the maximum character distance between the two knowledge entities in the target document, and Sim(kg1,kg2) refers to the cosine similarity between the two knowledge entities.

[0118] Then according to The relationship with a threshold (e.g., 0.3), such as when If the value is less than the threshold of 0.3, the relationship is removed, which means deleting the association between the current knowledge entity and other knowledge entities. This process is repeated for all adjacent knowledge entities to obtain the optimized target basic relationship structure.

[0119] Finally, based on the relationship between learners and knowledge entities, as well as the target-based relationship structure between knowledge entities, the final optimized target knowledge graph is obtained.

[0120] As an optional embodiment, a predictive model for community knowledge learning and knowledge production performance is constructed based on the feature values ​​of the three evolutionary modes in the above embodiments (i.e., "learner-mediated transitivity evolutionary mode", "knowledge-mediated transitivity evolutionary mode", and "learner reciprocity evolutionary mode").

[0121] If an individual or group's knowledge learning performance or knowledge production performance s can be obtained through a certain evaluation method, then a predictive model can be constructed with the evolutionary characteristics of the three modes as independent variables and knowledge learning performance or knowledge production performance as dependent variables.

[0122] Taking linear regression as an example:

[0123] s=β1F t1m1 +β2F t1m2 +β1F t1m3 +...+β 3n-2 F tnm1 +β 3n-1 F tnm2 +β 3n F tnm3 +β0

[0124] Among them, F tnm1 F tnm2 F tnm3 Let β0 represent the evolutionary eigenvalues ​​of "learner-mediated transitivity", "knowledge-mediated transitivity", and "learner reciprocity" at time tn, respectively, from β0 to β... 3n These are parameter values ​​obtained through iterative fitting during training.

[0125] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0126] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM (Read-Only Memory) / RAM (Random Access Memory), magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0127] According to another aspect of the embodiments of this application, an apparatus for knowledge evolution feature mining for implementing the above-described knowledge evolution feature mining method is also provided. Figure 7 This is a structural block diagram of an optional knowledge evolution feature mining apparatus according to an embodiment of this application, such as... Figure 7 As shown, the device may include:

[0128] The first acquisition module 701 is used to acquire the first relationship between multiple learners and multiple knowledge entities, the second relationship between multiple knowledge entities, and the third relationship between multiple learners at time t under the target knowledge graph.

[0129] The first determining module 702 is used to determine the evolutionary critical path obtained from the evolution process from time t to time t+1 based on the first relation, the second relation, the third relation and the target knowledge graph, wherein the evolutionary critical path is a critical path under multiple preset evolution modes.

[0130] The second determining module 703 is used to determine the target subject corresponding to each evolutionary critical path based on the evolutionary critical path;

[0131] The second acquisition module 704 is used to acquire the change trend values ​​of all evolutionary key paths associated with the target subject in the target knowledge graph. The change trend value represents the amount of change of the evolutionary key path, which is used to characterize whether the evolutionary characteristics corresponding to the evolutionary key path are valid.

[0132] The first module 705 is used to obtain the evolutionary characteristic value corresponding to the evolutionary characteristics by using the change trend value and the second-order relation coefficient of the target subject at time t.

[0133] It should be noted that the first acquisition module 701 in this embodiment can be used to execute the above step S201, the first determination module 702 in this embodiment can be used to execute the above step S202, the second determination module 703 in this embodiment can be used to execute the above step S203, the second acquisition module 704 in this embodiment can be used to execute the above step S204, and the first obtaining module 705 in this embodiment can be used to execute the above step S205.

[0134] Through the above modules, based on the relationships between learners, between learners and knowledge entities, and between knowledge entities at time t in the knowledge graph, the evolutionary critical path at time t+1 is obtained. Based on the evolutionary critical path, the corresponding target subject is determined, and then all trend values ​​associated with each target subject are obtained. Based on these trend values ​​and the second-order relational coefficients of the target subject at time t, the evolutionary characteristic values ​​of the evolutionary features corresponding to the evolutionary critical path are determined. After obtaining the evolutionary characteristic values, these values ​​can be used to support various prediction application scenarios in the knowledge graph-based online learning community, resulting in more accurate prediction results. This solves the problem that related technologies lack the mining of the community knowledge evolution feature in constructing community knowledge graphs, leading to inaccurate prediction results for various application scenarios within the knowledge graph-based online learning community.

[0135] As an optional embodiment, the second acquisition module includes:

[0136] The first determining unit is used to determine the preset calculation formula for relation extraction based on the evolutionary critical path;

[0137] The unit is used to obtain the trend value of each evolutionary critical path according to a preset calculation formula;

[0138] The first acquisition unit is used to acquire all trend values ​​corresponding to the target subject.

[0139] Optionally, the first determining unit includes:

[0140] The first setting submodule is used to extract the corresponding calculation formula of the learner's social network relationship as the preset calculation formula if the target subject at one end of the evolutionary critical path is the learner and the target subject at the other end is the learner.

[0141] The second setting submodule is used to extract the corresponding calculation formula of the relationship between learners and knowledge entities as a preset calculation formula if the target subject at one end of the evolution critical path is a learner and the other end is a knowledge entity.

[0142] Optionally, the first module includes:

[0143] The second acquisition unit is used to acquire the first relation value of the evolutionary critical path obtained by a preset calculation formula at time t.

[0144] The third acquisition unit is used to acquire the second relation value of the evolution critical path obtained by a preset calculation formula at time t+1.

[0145] The second determining unit is used to determine the trend value based on the first relationship value and the second relationship value.

[0146] Optionally, the second acquisition unit includes:

[0147] The first submodule is used to obtain the interaction behavior between a preset number of learners associated with the evolutionary key path at time t when the preset calculation formula is the corresponding calculation formula for extracting the learner's social network relationship, and to obtain the first interaction feature.

[0148] The second submodule is used to preprocess the first interaction feature to obtain the preprocessed second interaction feature;

[0149] The third submodule is used to obtain the first relationship value by utilizing the second interaction feature and the first weight;

[0150] or,

[0151] The fourth submodule is used to obtain the interaction behavior between learners and knowledge entities associated with the evolutionary key path at time t when the preset calculation formula is the corresponding calculation formula for the relationship between learners and knowledge entities, and obtain the third interaction feature.

[0152] The fifth submodule is used to preprocess the third interaction feature to obtain the preprocessed fourth interaction feature;

[0153] The sixth submodule is used to obtain the first relation value by utilizing the fourth interaction feature and the second weight.

[0154] Optionally, the third acquisition unit includes:

[0155] The seventh submodule is used to obtain the interaction behavior between a preset number of learners associated with the evolutionary key path at time t+1 when the preset calculation formula is the corresponding calculation formula for the learner's social network relationship extraction, and obtain the fifth interaction feature.

[0156] The eighth submodule is used to preprocess the fifth interaction feature to obtain the preprocessed sixth interaction feature;

[0157] The ninth submodule is used to obtain the second relation value by utilizing the sixth interaction feature and the first weight;

[0158] or,

[0159] The tenth submodule is used to obtain the interaction behavior between learners and knowledge entities associated with the evolutionary key path at time t+1 when the preset calculation formula is the corresponding calculation formula for the relationship between learners and knowledge entities. This results in the seventh interaction feature.

[0160] The eleventh submodule is used to preprocess the seventh interaction feature to obtain the preprocessed eighth interaction feature;

[0161] The twelfth submodule is used to obtain the second relation value by utilizing the eighth interaction feature and the second weight.

[0162] Optionally, the device further includes:

[0163] The third acquisition module is used to acquire the first relationship between multiple learners and multiple knowledge entities, the second relationship between multiple knowledge entities, and the third relationship between multiple learners at time t under the target knowledge graph. Before the target knowledge graph is composed of learners and knowledge entities, the module acquires the participants in the target scenario as learners.

[0164] The second module is used to identify and extract knowledge entities from the target text, and to construct hierarchical relationships between knowledge entities based on the target text to obtain the basic relationship structure.

[0165] The deletion module is used to delete the association between the preset number of knowledge entities to be optimized based on the average distance and corresponding cosine similarity between any two adjacent preset number of knowledge entities in the target text, so as to obtain the optimized target basic relationship structure.

[0166] The third module is used to obtain the target knowledge graph based on the relationship between learners and knowledge entities and the target basic relationship structure between knowledge entities.

[0167] It should be noted that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should also be noted that the above modules, as part of a device, can operate in environments such as... Figure 1 The hardware environment shown can be implemented through software or hardware, and the hardware environment includes the network environment.

[0168] According to another aspect of the embodiments of this application, an electronic device for implementing the above-described method for mining knowledge evolution features is also provided. The electronic device may be a server, a terminal, or a combination thereof.

[0169] Figure 8 This is a structural block diagram of an optional electronic device according to an embodiment of this application, such as... Figure 8 As shown, it includes a processor 801, a communication interface 802, a memory 803, and a communication bus 804. The processor 801, communication interface 802, and memory 803 communicate with each other via the communication bus 804.

[0170] Memory 803 is used to store computer programs;

[0171] When processor 801 executes a computer program stored in memory 803, it performs the following steps:

[0172] Obtain the first relationship between multiple learners and multiple knowledge entities, the second relationship between multiple knowledge entities, and the third relationship between multiple learners at time t under the target knowledge graph;

[0173] Based on the first relation, the second relation, the third relation, and the target knowledge graph, the evolutionary critical path obtained from the evolution process from time t to time t+1 is determined. The evolutionary critical path is the critical path under multiple preset evolutionary modes.

[0174] Based on the critical evolutionary paths, identify the target subject corresponding to each critical evolutionary path;

[0175] Obtain the change trend values ​​of all evolutionary critical paths associated with the target subject in the target knowledge graph. The change trend value represents the amount of change of the evolutionary critical path, which is used to characterize whether the evolutionary characteristics corresponding to the evolutionary critical path are valid.

[0176] By using the trend value and the second-order relation coefficient of the target subject at time t, the evolutionary characteristic value corresponding to the evolutionary characteristics is obtained.

[0177] Optionally, in this embodiment, the communication bus can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0178] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0179] The memory may include RAM, or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0180] As an example, such as Figure 8 As shown, the memory 803 may include, but is not limited to, the first acquisition module 701, the first determination module 702, the second determination module 703, the second acquisition module 704, and the first obtaining module 705 from the aforementioned knowledge evolution feature mining device. Furthermore, it may include, but is not limited to, other module units from the aforementioned knowledge evolution feature mining device, which will not be elaborated upon in this example.

[0181] The processors mentioned above can be general-purpose processors, including but not limited to: CPU (Central Processing Unit), NP (Network Processor), etc.; they can also be DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0182] In addition, the aforementioned electronic device also includes a display for showing the results of knowledge evolution feature mining.

[0183] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, and will not be repeated here.

[0184] Those skilled in the art will understand that Figure 8The structure shown is for illustrative purposes only. The device used to implement the above-mentioned knowledge evolution feature mining method can be a terminal device, such as a smartphone (e.g., Android phone, iOS phone), tablet computer, handheld computer, mobile internet device (MID), PAD, etc. Figure 8 This does not limit the structure of the aforementioned electronic devices. For example, the terminal device may also include components that are more advanced than those described above. Figure 8 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 8 The different configurations shown.

[0185] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, ROM, RAM, disk or optical disk, etc.

[0186] According to another aspect of the embodiments of this application, a storage medium is also provided. Optionally, in this embodiment, the storage medium can be used to execute program code for a method of knowledge evolution feature mining.

[0187] Optionally, in this embodiment, the storage medium may be located on at least one of the network devices in the network shown in the above embodiment.

[0188] Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps:

[0189] Obtain the first relationship between multiple learners and multiple knowledge entities, the second relationship between multiple knowledge entities, and the third relationship between multiple learners at time t under the target knowledge graph;

[0190] Based on the first relation, the second relation, the third relation, and the target knowledge graph, the evolutionary critical path obtained from the evolution process from time t to time t+1 is determined. The evolutionary critical path is the critical path under multiple preset evolutionary modes.

[0191] Based on the critical evolutionary paths, identify the target subject corresponding to each critical evolutionary path;

[0192] Obtain the change trend values ​​of all evolutionary critical paths associated with the target subject in the target knowledge graph. The change trend value represents the amount of change of the evolutionary critical path, which is used to characterize whether the evolutionary characteristics corresponding to the evolutionary critical path are valid.

[0193] By using the trend value and the second-order relation coefficient of the target subject at time t, the evolutionary characteristic value corresponding to the evolutionary characteristics is obtained.

[0194] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, and will not be repeated in this embodiment.

[0195] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, ROMs, RAMs, portable hard drives, magnetic disks, or optical disks.

[0196] According to another aspect of the embodiments of this application, a computer program product or computer program is also provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the computer device to perform the method steps of knowledge evolution feature mining in any of the above embodiments.

[0197] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0198] If the integrated units in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in the aforementioned computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the knowledge evolution feature mining method of the various embodiments of this application.

[0199] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0200] In the several embodiments provided in this application, it should be understood that the disclosed client can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or the indirect coupling or communication connection of units or modules may be electrical or other forms.

[0201] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the solution provided in this embodiment, depending on actual needs.

[0202] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or a predetermined number or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0203] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for mining knowledge evolution features, characterized in that, The method includes: Identify participants in the target scenario as learners; Knowledge entities are identified and extracted from the target text, and hierarchical relationships between the knowledge entities are constructed based on the target text to obtain the basic relationship structure; Based on the average distance and corresponding cosine similarity between any number of adjacent knowledge entities in the target text, the association between the number of knowledge entities to be optimized is deleted to obtain the optimized target basic relationship structure. Based on the association between the learner and the knowledge entity and the target basic relationship structure between the knowledge entities, a target knowledge graph is obtained, wherein the target knowledge graph is composed of the learner and the knowledge entity; Obtain the first relationship between multiple learners and multiple knowledge entities, the second relationship between multiple knowledge entities, and the third relationship between multiple learners at time t under the target knowledge graph; Based on the first relationship, the second relationship, the third relationship, and the target knowledge graph, the evolutionary critical path obtained from the evolution process from time t to time t+1 is determined, wherein the evolutionary critical path is a critical path under multiple preset evolutionary modes; Based on the aforementioned evolutionary critical paths, determine the target subject corresponding to each of the aforementioned evolutionary critical paths; Obtain the change trend values ​​of all the evolutionary key paths associated with the target subject in the target knowledge graph, wherein the change trend value represents the amount of change of the evolutionary key path, and is used to characterize whether the evolutionary characteristics corresponding to the evolutionary key path are valid; Using the trend value and the second-order relation coefficient of the target subject at time t, the evolutionary characteristic value corresponding to the evolutionary characteristics is obtained.

2. The method according to claim 1, characterized in that, The step of obtaining the change trend values ​​of all the evolutionary critical paths corresponding to the target subject in the target knowledge graph includes: The preset calculation formula for relation extraction is determined based on the aforementioned critical evolution path; The change trend value of each evolutionary critical path is obtained according to the preset calculation formula; Obtain all the trend values ​​corresponding to the target entity.

3. The method according to claim 2, characterized in that, The preset calculation formula for determining the relationship extraction based on the evolutionary critical path includes: If the target subject at one end of the evolutionary critical path is the learner and the learner is at the other end, then the calculation formula for extracting the learner's social network relationship is used as the preset calculation formula. If the target subject at one end of the evolutionary critical path is the learner and the knowledge entity at the other end is the knowledge entity, then the calculation formula corresponding to the relationship between the learner and the knowledge entity is extracted and used as the preset calculation formula.

4. The method according to claim 3, characterized in that, The step of obtaining the change trend value of each evolutionary critical path according to the preset calculation formula includes: Obtain the first relation value of the evolutionary critical path at time t, calculated using the preset formula; Obtain the second relation value of the evolutionary critical path at time t+1, calculated using the preset formula; The trend value is determined based on the first relationship value and the second relationship value.

5. The method according to claim 4, characterized in that, The step of obtaining the first relation value of the evolutionary critical path at time t, obtained by the preset calculation formula, includes: When the preset calculation formula is the calculation formula corresponding to the learner's social network relationship extraction, the interaction behavior between a preset number of learners associated with the evolutionary key path at time t is obtained to obtain the first interaction feature; The first interaction feature is preprocessed to obtain the preprocessed second interaction feature; The first relationship value is obtained by using the second interaction feature and the first weight; or, When the preset calculation formula is the calculation formula corresponding to the relationship extraction between learners and knowledge entities, the interaction behavior between the learner and the knowledge entity associated with the evolutionary key path at time t is obtained to obtain the third interaction feature; The third interaction feature is preprocessed to obtain the preprocessed fourth interaction feature; The first relationship value is obtained by utilizing the fourth interaction feature and the second weight.

6. The method according to claim 4, characterized in that, The step of obtaining the second relation value of the evolutionary critical path at time t+1, calculated by the preset formula, includes: When the preset calculation formula is the calculation formula corresponding to the learner's social network relationship extraction, the interaction behavior between a preset number of learners associated with the evolutionary key path at the (t+1)th time is obtained to obtain the fifth interaction feature; The fifth interaction feature is preprocessed to obtain the preprocessed sixth interaction feature; The second relationship value is obtained by using the sixth interaction feature and the first weight; or, When the preset calculation formula is the calculation formula corresponding to the relationship extraction between learners and knowledge entities, the interaction behavior between the learner and the knowledge entity associated with the evolutionary key path at the (t+1)th time is obtained to obtain the seventh interaction feature; The seventh interaction feature is preprocessed to obtain the preprocessed eighth interaction feature; The second relationship value is obtained by utilizing the eighth interaction feature and the second weight.

7. An apparatus for mining knowledge evolution features, characterized in that, The device includes: The third acquisition module is used to acquire participants in the target scenario as learners; The second module is used to identify and extract knowledge entities from the target text, and construct hierarchical relationships between the knowledge entities based on the target text to obtain a basic relationship structure. The deletion module is used to delete the association between the preset number of knowledge entities to be optimized based on the average distance and corresponding cosine similarity between any two adjacent preset number of knowledge entities in the target text, so as to obtain the optimized target basic relationship structure. The third obtaining module is used to obtain a target knowledge graph based on the association between the learner and the knowledge entity and the target basic relationship structure between the knowledge entities, wherein the target knowledge graph is composed of the learner and the knowledge entity; the first obtaining module is used to obtain the first relationship between multiple learners and multiple knowledge entities, the second relationship between multiple knowledge entities, and the third relationship between multiple learners at time t under the target knowledge graph. The first determining module is used to determine the evolutionary critical path obtained from the evolution process from time t to time t+1 based on the first relationship, the second relationship, the third relationship and the target knowledge graph, wherein the evolutionary critical path is a critical path under multiple preset evolution modes; The second determining module is used to determine the target subject corresponding to each evolutionary critical path based on the evolutionary critical path; The second acquisition module is used to acquire the change trend values ​​of all the evolutionary key paths associated with the target subject in the target knowledge graph, wherein the change trend value represents the change amount of the evolutionary key path, and is used to characterize whether the evolutionary characteristics corresponding to the evolutionary key path are valid. The first obtaining module is used to obtain the evolutionary characteristic value corresponding to the evolutionary characteristics by using the change trend value and the second-order relation coefficient of the target subject at time t.

8. An electronic device comprising a processor, a communication interface, a memory, and a communication bus, wherein, The processor, the communication interface, and the memory communicate with each other via the communication bus, characterized in that... The memory is used to store computer programs; The processor is configured to perform the method steps of any one of claims 1 to 6 by running the computer program stored in the memory.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the method described in any one of claims 1 to 6.

Images