Future industry evolution mechanism prediction method and device based on time sequence hypergraph motif

Abstract

The invention discloses a future industry evolution mechanism prediction method and device based on a time sequence hypergraph motif, and belongs to the technical field of big data intelligent processing, and the method comprises the steps: collecting an original cooperation data set among a plurality of mechanisms in a target industry field, and generating a cooperation event data set containing a timestamp; constructing a time sequence cooperation hypergraph; determining the real number of pre-selected P types of target time sequence hypergraph motifs in the time sequence cooperation hypergraph, wherein P is greater than or equal to 1; creating a zero-model random hypergraph set, and counting the average number of each type of target time sequence hypergraph motifs in the zero-model random hypergraph set; calculating the abundance of each type of target time sequence hypergraph motifs according to the real number and the average number; and constructing a feature profile of the time sequence cooperation hypergraph based on abundance, and analyzing the change trend of the feature profile to predict the evolution mechanism of the target industry. According to the invention, the evolution mechanism of the future industry can be analyzed more meticulously, and the interaction mode among the plurality of innovation subjects in the whole industry can be visually represented.

Description

Technical Field

[0001] The present invention relates to the field of big data intelligent processing technology, and in particular to a method and device for predicting future industrial evolution mechanisms based on a time-series hypergraph motif. Background Art

[0002] With the rapid development of science and technology, future industries represented by quantum information technology, brain-computer interfaces, and humanoid robots are becoming strategic commanding heights in global science and technology competition. Such industries are usually spawned by the industrialization of major scientific and technological innovations, and have three major characteristics: disruptive technological breakthroughs, industrial ecological reconstruction, and strategic value leadership. However, the development of these industries also faces many challenges, such as high uncertainty in technology routes, strong heterogeneity of innovation entities, and large resource requirements and complex configurations.

[0003] Specifically, existing research and methods have the following deficiencies in analyzing future industrial collaborative innovation models:

[0004] 1. Ignoring the temporal hypergraph structure. The development of future industries is highly dependent on breakthroughs in future technologies, and the research and development of these technologies often requires the participation of multiple institutions to form complex cluster cooperation. The classic network structure may not be able to capture the emergent behavior caused by cluster cooperation, while the temporal hypergraph structure can more accurately model the nonlinearity and temporal nature of institutional cooperation in future industries. However, the existing temporal hypergraph construction methods have limitations. For example, the Motif-based temporal academic hypergraph network reasoning method proposed in Chinese patent document CN117350385A, whose temporal hypergraph is constructed in the form of discrete snapshots, ignores the connections between different snapshots.

[0005] 2. Ignoring the role of network motifs. Motifs are local topological primitives of networks and play an important role in promoting the evolution of networks (reference: Science (2002), 298 (5594), 824-827p). The existing technology has two limitations. One is that it emphasizes the global structure. For example, Chinese patent document CN118656713A predicts the future direction of industries by analyzing the weighted degree centrality of industrial cluster networks. However, networks with similar global structures may have different local structures, and networks with different global structures may also present similar local structures (reference: Science (2004), 303 (5663), 1538-1542). Network motifs provide a good entry point for studying local structures. Compared with the macroscopic global perspective, network motifs can study the interactive relationship between institutions in a more detailed way at the mesoscale level. The second is to ignore the hypergraph structure. For example, Chinese patent document CN115766476A proposed a network similarity evaluation method based on network motifs, but did not introduce a temporal hypergraph structure to model the institutional cooperation network in future industries. Summary of the invention

[0006] The purpose of the present invention is to provide a method and device for predicting the future industry evolution mechanism based on a temporal hypergraph motif, so as to solve the problem that the prior art cannot analyze the evolution mechanism of future industries and the interaction mode between multiple innovation subjects in detail when analyzing the future industry collaborative innovation model because it ignores the role of the temporal hypergraph structure and network motifs.

[0007] To achieve the above objectives, this application adopts the following technical solutions:

[0008] In a first aspect of the present application, a method for predicting future industrial evolution mechanisms based on a temporal hypergraph motif is provided, comprising the following steps:

[0009] Collecting original cooperation data sets between several institutions in the target industry field, and preprocessing the original cooperation data sets to generate cooperation event data sets containing timestamps;

[0010] Constructing a temporal cooperation hypergraph according to the cooperation event dataset, wherein nodes represent organizations, hyperedges represent cooperation events, and are marked with corresponding timestamps;

[0011] Determine the actual number of P-type target temporal hypergraph motifs pre-selected in the temporal cooperation hypergraph by using a counting algorithm, P ≥ 1;

[0012] Creating a zero model random hypergraph set, and counting the average number of each type of target time series hypergraph motifs in the zero model random hypergraph set;

[0013] Calculate the abundance of each type of target temporal hypergraph motif according to its true number in the temporal cooperation hypergraph and its average number in the zero model random hypergraph set;

[0014] Based on the abundance of all target temporal hypergraph motifs, a characteristic profile of the temporal cooperation hypergraph is constructed, and the changing trend of the characteristic profile is analyzed to predict the evolution mechanism of the target industry.

[0015] In a second aspect of the present application, a device for predicting future industrial evolution mechanisms based on a temporal hypergraph motif is provided, comprising:

[0016] A preprocessing module is used to collect the original cooperation data set between several institutions in the target industry field, and preprocess the original cooperation data set to generate a cooperation event data set containing a timestamp;

[0017] A hypergraph construction module, used to construct a time-series cooperation hypergraph based on the cooperation event dataset, wherein nodes represent organizations, hyperedges represent cooperation events, and are marked with corresponding timestamps;

[0018] A motif detection module, used for determining the actual number of P-type target temporal hypergraph motifs pre-selected in the temporal cooperation hypergraph by using a counting algorithm, where P≥1;

[0019] A statistical module, used to create a zero model random hypergraph set, and count the average number of each type of target time series hypergraph motifs in the zero model random hypergraph set;

[0020] An abundance calculation module is used to calculate the abundance of each type of target time-series hypergraph motif according to its actual number in the time-series cooperation hypergraph and its average number in the zero model random hypergraph set;

[0021] A prediction module is used to construct a characteristic profile of the temporal cooperation hypergraph based on the abundance of all target temporal hypergraph motifs, and analyze the changing trend of the characteristic profile to predict the evolution mechanism of the target industry.

[0022] The third aspect of the present application provides an electronic device comprising a memory and a processor, wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement a method for predicting future industrial evolution mechanisms based on a time-series hypergraph motif as described in any one of the above.

[0023] The fourth aspect of the present application provides a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to implement a method for predicting future industrial evolution mechanisms based on a time-series hypergraph motif as described above when executed.

[0024] The present invention has the following beneficial effects:

[0025] 1. By introducing the temporal hypergraph structure, the assumption of pairwise interaction of nodes in the classical network structure is extended to the high-order interaction between multiple nodes, and the timing and dynamics of these high-order interactions are taken into account, which can more carefully analyze the evolution mechanism of future industries.

[0026] 2. By introducing network motifs, the future industrial cooperation network is deconstructed from the mesoscopic level. Compared with the previous analysis of the macroscopic global structure, the motif-based analysis can consider the impact of specific types of local topological structures on the network formation mechanism and network dynamics; at the same time, the motif, as a special subgraph, can also intuitively represent the interaction pattern between multiple innovation entities in the entire industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0028] Figure 1 It is a flow chart of a method for predicting future industrial evolution mechanism based on a time-series hypergraph motif provided in an embodiment of the present application;

[0029] Figure 2 It is a schematic diagram of the structure of a device for predicting future industrial evolution mechanism based on a time-series hypergraph motif provided in an embodiment of the present application;

[0030] Figure 3 It is a schematic diagram of an electronic device for implementing a method for predicting future industrial evolution mechanisms based on a time-series hypergraph motif provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] To make the technical solution of the present application clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The terms "first", "second", etc. in the claims and specification of the present application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances. This is only to describe the distinction method used when describing objects with the same attributes in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that a process, method, system, product or device containing a series of units is not necessarily limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or devices.

[0032] This embodiment provides a method for predicting the future industrial evolution mechanism based on a time series hypergraph motif. Figure 1 As shown, the following steps are included:

[0033] S110, collecting original cooperation data sets between several institutions in the target industry field, and preprocessing the original cooperation data sets to generate cooperation event data sets containing timestamps.

[0034] Among them, the original cooperative data sets among several institutions in the target industry are collected, including:

[0035] Determine the target industry and extract the keywords corresponding to the target industry;

[0036] Use field keywords as search conditions to obtain a list of institutions related to the target industry;

[0037] Based on the list of institutions, we collect cooperation events between institutions to form an original cooperation data set. The first step in data collection is to determine the future industry that needs to be investigated, that is, the target industry, and then obtain data in the target industry that can characterize the cooperation relationship between different institutions to form the original cooperation data set. Obviously, the cooperation relationship between institutions will affect the data collection results. Therefore, it is also necessary to determine the representation form of cluster cooperation relationship. In this embodiment, the representation form of cooperation relationship between institutions includes but is not limited to jointly publishing a paper, jointly applying for a patent, jointly writing a book, or jointly participating in an industry alliance.

[0038] Taking the representation form of co-authorship as an example, after determining the target industry and the form of cooperation, we first rely on the knowledge of domain experts to determine the domain keywords of the target industry. Then, based on the domain keywords, we retrieve relevant papers through public data platforms such as OpenAlex, Dimensions, Web of Science, Scopus, and PubMed, and only retain specific types of documents to form the original cooperation data set, including journal articles, conference papers, books, and book chapters. The data collection method and the retained document types here are related to the form of cooperation between institutions.

[0039] After completing the search, key information of the paper was screened, such as title, abstract, journal name (conference name), author, signing institution, publication year and Digital Object Identifier (DOI), etc., to pre-process the original collaborative dataset. DOI is a standardized coding system for uniquely and permanently identifying digital resources, similar to a "digital ID card", which aims to provide a globally unique and resolvable identifier for various objects in the digital environment (such as academic papers, e-books, research reports, datasets, etc.).

[0040] In other forms of cooperation, such as supply chain networks, production cooperation networks, and mergers and acquisitions networks, after determining the target industry, semantic analysis technology can be used to conduct in-depth analysis of relevant literature on the target industry, accurately extract the field keywords corresponding to the target industry, and then use network search engines or professional industry databases to use field keywords as search conditions to obtain a list of institutions closely related to the target industry. Based on the list of institutions, through multi-channel information collection, including news reports, corporate announcements, and data mining technology, we can comprehensively obtain cooperation events between various institutions and form an original cooperation data set.

[0041] The original cooperation data set is preprocessed to generate a cooperation event data set containing timestamps, including:

[0042] Disambiguate the names of institutions to ensure that the same institution has a unified name in all cooperation events;

[0043] Assign a corresponding timestamp to each cooperation event according to the preset time unit;

[0044] When multiple cooperation events have the same timestamp, unique timestamps are configured for the multiple cooperation events by random sorting.

[0045] The first step in data preprocessing is to disambiguate the institution names. In the original cooperation data set, the same institution may correspond to multiple institution names, such as inconsistent abbreviations, former names, and incorrect signatures. At the same time, sub-institutions at different levels in the same institution may also sign differently, which can easily lead to data confusion and difficulty in analysis. Therefore, in data preprocessing, it is necessary to disambiguate the institution names. Specifically, first identify all the institution name variants in the original cooperation data set, and then select a standard name for each institution. If the institution has a hierarchy, its sub-institutions are merged into the same hierarchy, thereby ensuring that the same institution mentioned in all cooperation events uses the same name, which is convenient for subsequent data analysis and visualization.

[0046] It should be noted here that, in theory, there must be a signed institution in the paper. Therefore, before disambiguating the institution name, it is necessary to screen out the data with missing signed institutions, and then re-search in other public databases. If the missing institution information is obtained, it will be used to fill the missing value. If it cannot be obtained, search for authors with the same name in the collected data and use the signed institutions of the author in other data that are not missing to fill it. If it still cannot be obtained, delete this data. In other words, supplementing the missing signed institutions is not a necessary step in data preprocessing, but is selected according to the actual situation.

[0047] Finally, assign a corresponding timestamp to each paper according to the year. Specifically, different time units such as days, weeks, months, years and corresponding time windows can be determined according to the frequency and degree of change of cooperative behaviors between institutions. However, in any case, it is necessary to ensure that each set of cooperative relationships has its corresponding timestamp. When the time unit is larger, such as a year, there may be multiple cooperative relationships with the same timestamp. At this time, a random sorting method is adopted to sort each identical timestamp and configure a unique timestamp. For example, the timestamps of cooperative events A, B and C are all 2025, then the randomly assigned timestamps can be: cooperative event A corresponds to April 5, 2025; cooperative event B corresponds to April 6, 2025; cooperative event C corresponds to April 7, 2025.

[0048] Preprocessing operations such as institution name disambiguation and timestamp assignment significantly improved the quality and consistency of the data and provided a solid foundation for subsequent data analysis.

[0049] S120. Construct a temporal cooperation hypergraph based on the cooperation event dataset, wherein nodes represent organizations, hyperedges represent cooperation events, and are marked with corresponding timestamps.

[0050] Define a hypergraph based on the cluster collaboration between institutions ,in, Represents a set of nodes, each node represents an organization, represents a set of hyperedges, and , Each element in Represents a hyperedge, mathematically represented as a non-empty subset containing any number of nodes, in the node set Temporal Hypergraph , represents an ordered sequence of temporal hyperedges. In this embodiment, the temporal hypergraph is also called a temporal cooperation hypergraph, which represents the cooperation relationship between organizations. In the temporal cooperation hypergraph, each temporal hyperedge is represented by , which contains the node set , is the timestamp corresponding to the temporal hyperedge, that is, the time point when the cooperation event occurred. At the same time, the timestamp of each temporal hyperedge is unique. Taking the co-authorship relationship as an example, a paper is regarded as a hyperedge formed by its publishing institution. The number of nodes contained in the hyperedge is the number of publishing institutions of the paper. After adding the year as the timestamp, each hyperedge becomes a temporal hyperedge. When two timing superedges and It is considered to be a duplicate. The definition contains the node set The temporal hyperedge set of , when timestamps and repeated temporal hyperedges are ignored, the temporal hypergraph Static Hypergraph , if and only if hour, .

[0051] Assume that in 2021, Institutions 1 and 2 jointly published Paper 1, in 2022, Institutions 2, 3, and 4 jointly published Paper 2, in 2023, Institutions 1 and 3 jointly published Paper 3, in 2024, Institutions 4 and 5 jointly invented Paper 4, and in 2025, Institutions 1, 2, 3, and 4 jointly invented Paper 5. Then, in the constructed temporal collaboration hypergraph, In the node set ,in, They correspond to institutions 1, 2, 3, 4 and 5 respectively, corresponding to the hyperedge of paper 1 , timestamp is 2021; corresponding to the hyperedge of Paper 2 , timestamp is 2022; corresponding to the hyperedge in Paper 3 , timestamp is 2023; corresponding to the hyperedge in Paper 4 , timestamp is 2024; corresponding to the hyperedge in paper 5 , with a timestamp of 2025.

[0052] The classic network structure can only model pairwise interactions between nodes, while the hypergraph can incorporate a wider range of group interactions into the model. The time-series hypergraph further considers the dynamics of such group interactions and extends the assumption of pairwise interactions of nodes in the classic network structure to high-order interactions between multiple nodes, enabling a more detailed analysis of the evolution mechanism of future industries in the future.

[0053] In this embodiment, the time-series cooperation hypergraph can be used to analyze how the cooperation patterns among different innovation entities evolve over time, and can also identify core institutions or key cooperation patterns.

[0054] S130. Determine the actual number of pre-selected P-type target temporal hypergraph motifs in the temporal cooperative hypergraph using a counting algorithm, where P≥1.

[0055] After obtaining the temporal cooperative hypergraph, the actual number of each type of target temporal hypergraph motif (TH-motif) in the temporal cooperative hypergraph can be determined by the precise counting algorithm THyMe+ or the approximate counting algorithm THyMe-A*.

[0056] It should be noted here that due to the complexity of the hypergraph structure, after considering the temporal nature, only the temporal hypergraph motif formed by three temporal hyperedges is detected. The temporal hypergraph motif describes the The structure and temporal relationship presented by the sequence formed by three connected temporal hyperedges.

[0057] When considering the temporal hypergraph motifs formed by three temporal hyperedges, after excluding redundancy, there are a total of 96 motifs containing three temporal hyperedges. Given three connected temporal hyperedges ,in ,and , these motifs can be divided into three categories according to the number of repeated temporal hyperedges they contain: (1) ternary TH-motif, which contains three non-repeating temporal hyperedges, i.e. ; (2) binary TH-motif, which contains two repeated temporal hyperedges, e.g. ; (3) Unary TH-motif, containing three repeated temporal hyperedges, namely In this embodiment, the ternary temporal hypergraph motif, the binary temporal hypergraph motif and the unary temporal hypergraph motif are all target temporal hypergraph motifs, that is, P=3.

[0058] Then, the number of motifs is detected.

[0059] Among them, when the counting algorithm adopts the precise counting algorithm based on temporal hypergraph projection, namely THyMe+, the counting algorithm is used to determine the actual number of P-type target temporal hypergraph motifs pre-selected in the temporal cooperation hypergraph, P≥1, including:

[0060] Given a time window δ, and initialize the projection map , where node v∈ The set of hyperedges with the same node set in the corresponding time-series cooperation hypergraph , the timestamp set carried by each node v = ,side ∈ If and only if ∩ ≠ .

[0061] Update based on inserting modules and deleting modules :

[0062] In the insertion module, given any timing hyperedge , if there is no node set with it in the current time window The same hyperedge, the corresponding node is added as a new node and insert Then, find its adjacent nodes , that is, in the temporal cooperative hypergraph Find Temporal hyperedges with the same nodes, then create associated edges Finally, The corresponding timestamp Add to middle.

[0063] In the Delete module, if the selected timing hyperedge is the current time window The only node set in The timing hyperedge , then put it in The nodes and associated edges represented in and Before deleting it, Timestamp from Delete it.

[0064] When the projection After updating, start counting steps. Given any temporal hyperedge , THyMe+ algorithm in the time window Including The motifs are counted, where is the last hyperedge in the motif, and THyMe+ is in the time window Find any two sums Connected temporal hyperedges.

[0065] Next, the number of ternary temporal hypergraph motifs, binary temporal hypergraph motifs, and uniary temporal hypergraph motifs are calculated respectively.

[0066] First calculate the number of ternary temporal hypergraph motifs, given three connected temporal hyperedge sets ,because is the last temporal hyperedge to appear, so the set There are two sequences or Get, where , ,because , ,and , then the number of motifs can be obtained by and It is calculated by the number of combinations of timestamps contained in . represents the number of ternary temporal hypergraph motifs, then , ,in, stands for indicator function. and Indicates when is the last occurring temporal hyperedge, the temporal hyperedge set Considering and There are two possible temporal hypergraph motifs that may be formed after the timing.

[0067] Then we count the number of binary temporal hypergraph motifs that have two repeated hyperedges, which is equivalent to Enumerate all Adjacent edges As with the calculation of the ternary temporal hypergraph motif, consider three sequences , and ,in , ,because , , the number of motifs can also be obtained by calculating the number of combinations of timestamps it contains, that is, , as well as .

[0068] Finally, the number of univariate temporal hypergraph motifs is calculated, given the sequence ,in , .because , ,and ,at this time .

[0069] Assume that in the time-series cooperative hypergraph In the node set , A, B and C represent organizations, and the set of time-series hyperedges Include , , , , and set the time window δ=3.

[0070] Initially, the projection is empty.

[0071] Insert module inserts timing hyperedges one by one: Insert , check the current time window Is there already an identical set of nodes in , there are currently no other hyperedges, so Add to and update , , , ;insert , check the current time window Is there already an identical node set in , there are currently no other hyperedges, so Add to and update , , , , ;insert , check the current time window Is there already an identical set of nodes in , there are currently no other hyperedges, so Add to and update , , , , , ; Because δ=3, we do not insert .

[0072] The delete module checks the hyperedges outside the time window: When time advances, if the timestamp of a hyperedge exceeds the current time window, it will be removed from the Take the current time as t = 5 as an example, then the time window is , super edge The timestamp of is not within the window, so delete it and update , , , , , In the time window In, including Count the motifs.

[0073] by Take this as an example, first calculate the ternary temporal hyperedge motif:

[0074] and All adjacent hyperedges are and , calculate the possible sequences , whose timestamp combination =2, =3, =4, ; and possible sequences , whose timestamp combination =3, =2, =4, .

[0075] Then calculate the binary temporal hyperedge motif: check the repeated hyperedges as follows and , calculate the number of timestamp combinations .

[0076] Finally, the univariate temporal hyperedge motif is calculated: for a single hyperedge , the number of timestamp combinations = .

[0077] For real hypergraphs, when the hypergraph is large in size, that is, the number of hyperedges is large, or the time window spanned is long, the accurate counting algorithm often takes a long time and requires higher computing resources. At this time, based on the idea of ​​random sampling, the THyMe-A* algorithm can be used to obtain an unbiased estimate of the number of motifs in each type of temporal hypergraph. The core of the THyMe-A* algorithm is to repeat the sampling length time window, and then accurately count the number of temporal hypergraph motifs in each time window according to the THyMe+ algorithm, and finally obtain an unbiased estimate of the number of temporal hypergraph motifs in the entire time window.

[0078] Specifically, when the counting algorithm adopts an approximate counting method based on random sampling, the counting algorithm is used to determine the true number of P-type target temporal hypergraph motifs pre-selected in the temporal cooperation hypergraph, P ≥ 1, including:

[0079] Set the sampling time window length ∆ and the number of sampling times s.

[0080] For each sampling, the following steps are performed:

[0081] a. From the temporal hyperedge set Randomly and uniformly select a temporal hyperedge ;

[0082] b. According to the selected timing hyperedge The timestamp t, from the time window [ ]A random timestamp is selected ;

[0083] c. Based on timestamp , using the THyMe+ algorithm to calculate the formation time in the sub-window [ ] to accurately count each target temporal hypergraph motif and obtain ;

[0084] d. Calculate the timing hyperedge that appears in the time window [ The probability of ,in, is the number of hyperedges in the subwindow;

[0085] e.According to Adjustment ,get .

[0086] The estimated number of temporal hypergraph motifs of each type obtained in each sampling Add the total and then multiply by , we obtain the final unbiased estimator of the number of each type of temporal hypergraph motifs, which is also regarded as the true number of each type of target temporal hypergraph motifs.

[0087] Taking the above time-series cooperation hypergraph as an example, assuming that the time window , and set the sampling times s=10, randomly and uniformly select a time series hyperedge ,for , whose timestamp , so the time window is [ ]=[ ], and then randomly select a timestamp , then the subwindow is [ ], then the above-mentioned THyMe+ algorithm needs to be used to calculate the number of all possible temporal hypergraph motifs in this sub-window.

[0088] Taking the calculation of the number of ternary temporal hypergraph motifs as an example, check and In this subwindow [ ], calculate the number of combinations (because ),Right now . Because in the child window [ ] contains , ,so , , and then calculate .

[0089] After 10 samplings, the results of each sampling are similar, such as , then the total estimate is , that is, in this embodiment, the actual number of ternary temporal hypergraph motifs is 2.

[0090] Deconstructing the future industrial cooperation network from a mesoscopic level, compared with previous analyses of macroscopic global structures, motif-based analysis can consider the impact of specific types of local topological structures on network formation mechanisms and network dynamics; at the same time, motifs, as a special subgraph, can intuitively represent the interaction patterns between multiple innovation entities in the entire industry.

[0091] In this embodiment, by analyzing which nodes frequently appear in various models, institutions or organizations that play a key role in the industrial development process can be identified. For example, in the field of science and technology, certain research institutions may play an important role in promoting technological progress due to their extensive interdisciplinary cooperation.

[0092] S140, creating a zero model random hypergraph set, and counting the average number of each type of target time series hypergraph motifs in the zero model random hypergraph set.

[0093] Further, a set of zero-model random hypergraphs are constructed using a zero-model generation method, wherein each random hypergraph retains the basic statistical properties of the temporal cooperation hypergraph;

[0094] According to the above counting method, the number of occurrences of each type of target temporal hypergraph motif in each random hypergraph is calculated, and the average value is recorded.

[0095] Using the Chung-Lu model or other appropriate methods, a set of zero-model random hypergraphs are generated. Each set of hypergraphs retains the basic statistical properties of the original temporal cooperation hypergraph (such as degree distribution), but other aspects are randomized.

[0096] In each zero-model random hypergraph, the number of occurrences of each type of target temporal hypergraph motif is calculated using the above-mentioned exact counting algorithm THyMe+ or approximate counting algorithm THyMe-A*, and then the average number of each type of target temporal hypergraph motif in all zero-model random hypergraphs is calculated.

[0097] S150. Calculate the abundance of each type of target temporal hypergraph motif based on its actual number in the temporal cooperation hypergraph and its average number in the zero model random hypergraph set.

[0098] By formula Calculate the abundance of each type of temporal hypergraph motif, where express Medium temporal hypergraph motif The real number of Represents the temporal hypergraph motifs generated in all random networks The mean of An adjustment factor (usually set to 1) to avoid zero denominator.

[0099] In this embodiment, a pre-set p-value can also be used to determine whether each type of target temporal hypergraph motif is statistically significant, thereby identifying which cooperation patterns in the actual network are significantly different from random expectations. This method not only helps to understand the structural characteristics of complex networks, but also provides data-driven support for policy making.

[0100] Assuming that motif type 1 and type 2 are significantly more common in the actual network than expected by chance, it indicates that these two cooperation modes may be important characteristics of the industry. For example, if motif type 1 is a "strong core" structure, the institutions participating in the cooperation and their cooperation relationships remain unchanged within a given time window, indicating that there is a stable industrial cluster in the industry; if motif type 2 is a "core-periphery" structure, some institutions continue to maintain a stable cooperation relationship within a given time window, while there are some temporary institutions participating in the cooperation in the remaining stages, indicating that there are stable core institutions in the industry, and there are also peripheral institutions playing a supporting role; although motif type 3 is also statistically significant, its p value is close to the threshold, indicating that the importance of this cooperation mode is relatively low. Based on this analysis, for cooperation modes that are significantly more common than expected by chance, such as the "strong core" structure and the "core-periphery" structure, these modes can be further strengthened through policy support such as funding cross-institutional projects, and resources can be preferentially tilted to institutions that show significant cooperation modes to maximize innovation output.

[0101] S160. Construct a characteristic profile of the temporal cooperation hypergraph based on the abundance of all target temporal hypergraph motifs, and analyze the changing trend of the characteristic profile to predict the evolution mechanism of the target industry.

[0102] Further, the normalized abundance of each type of target temporal hypergraph motif is calculated, and the normalized abundance is used as a dimension to construct a characteristic profile of the temporal cooperation hypergraph;

[0103] Analyze the differences in the characteristic profiles in different periods and predict the evolution mechanism of the target industry.

[0104] In view of the fact that there may be comparisons between different hypergraphs based on temporal hypergraph motifs, such as cross-industry comparisons or international comparisons of the same industry, it is necessary to construct a feature profile of a single temporal cooperation hypergraph. In this embodiment, for each temporal cooperation hypergraph, since there are three different target temporal hypergraph motifs, the feature profile of each temporal hypergraph can be represented by a three-dimensional vector, where each dimension represents the normalized abundance of a type of target temporal hypergraph motif. ,This normalization process ensures that networks of different sizes can be compared fairly.

[0105] Conduct in-depth analysis on the constructed characteristic profiles, including but not limited to time series analysis, pattern recognition, and correlation analysis with other variables, in order to understand the changing trends of characteristic profiles in different periods, and reveal some potential laws or trends by analyzing the changes in characteristic profiles over time, and then predict the future evolution direction or mechanism of the target industry. For example, certain types of cooperation models may herald the arrival of a technological innovation cycle or major changes in the market structure, thereby providing reference opinions for policy making.

[0106] For example, the analysis found that the characteristic profiles before and after 2018 showed obvious differences. Before 2018, the network was mainly dominated by a few large institutions, forming a highly centralized cooperation model; after 2018, there was more extensive cooperation between small and medium-sized institutions, showing a more decentralized but active cooperation ecology. This shift may indicate a major change in the market structure of the industry, from monopoly to diversified competition. Based on this, policymakers focus on the development needs of small and medium-sized enterprises and create a fair competition environment for them to promote the healthy and stable development of the market.

[0107] like Figure 2 As shown, this embodiment also provides a device for predicting future industrial evolution mechanism based on a time series hypergraph motif, comprising:

[0108] The preprocessing module is used to collect the original cooperation data set between several institutions in the target industry field, and preprocess the original cooperation data set to generate a cooperation event data set containing timestamps;

[0109] A hypergraph construction module is used to construct a temporal cooperation hypergraph based on the cooperation event dataset, where nodes represent organizations and hyperedges represent cooperation events with corresponding timestamps.

[0110] A motif detection module is used to determine the actual number of P-type target temporal hypergraph motifs pre-selected in the temporal cooperation hypergraph by using a counting algorithm, where P≥1;

[0111] A statistical module is used to create a zero model random hypergraph set and count the average number of each type of target time series hypergraph motifs in the zero model random hypergraph set;

[0112] An abundance calculation module, used to calculate the abundance of each type of target temporal hypergraph motif according to its actual number in the temporal cooperation hypergraph and its average number in the zero model random hypergraph set;

[0113] The prediction module is used to construct the characteristic profile of the temporal cooperation hypergraph based on the abundance of all target temporal hypergraph motifs, and analyze the changing trend of the characteristic profile to predict the evolution mechanism of the target industry.

[0114] This embodiment is used to implement the method provided by the above embodiment, and has the corresponding beneficial effects of the above method. For technical details not fully described in this embodiment, please refer to the methods provided by all the above embodiments of the present invention.

[0115] like Figure 3 As shown, this embodiment also provides an electronic device, including a memory 301 and a processor 302, wherein the memory 301 is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor 302 to implement the above-mentioned method for predicting the future industrial evolution mechanism based on a time-series hypergraph motif.

[0116] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the electronic device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0117] This embodiment also provides a computer-readable storage medium storing a computer program, and the computer program enables the computer to implement the above-mentioned method for predicting the future industrial evolution mechanism based on the time-series hypergraph model when executed.

[0118] Exemplarily, the computer program may be divided into one or more modules / units, one or more modules / units are stored in the memory 301 and executed by the processor 302, and the input interface 305 and the output interface 306 complete the I / O interface transmission of data to complete the present invention. The one or more modules / units may be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the computer device.

[0119] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may include, but is not limited to, a memory 301 and a processor 302. Those skilled in the art may understand that this embodiment is only an example of a computer device and does not constitute a limitation on the computer device. The computer device may include more or fewer components, or a combination of certain components, or different components. For example, the computer device may also include an input device 307, a network access device, a bus, etc.

[0120] The processor 302 may be a central processing unit (CPU), or other general-purpose processors 302, digital signal processors 302 (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor 302 may be a microprocessor 302, or the processor 302 may be any conventional processor 302, etc.

[0121] The memory 301 may be an internal storage unit of a computer device, such as a hard disk or memory of the computer device. The memory 301 may also be an external storage device of the computer device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device. Furthermore, the memory 301 may include both an internal storage unit of the computer device and an external storage device. The memory 301 is used to store computer programs and other programs and data required by the computer device. The memory 301 may also be used to temporarily store in the output device 308. The aforementioned storage media include various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory ROM 303, a random access memory RAM 304, a disk or an optical disk.

[0122] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for predicting future industrial evolution mechanisms based on a temporal hypergraph motif, characterized in that: The following steps are involved: Collecting original cooperation data sets between several institutions in the target industry field, and preprocessing the original cooperation data sets to generate cooperation event data sets containing timestamps; Constructing a temporal cooperation hypergraph according to the cooperation event dataset, wherein nodes represent organizations, hyperedges represent cooperation events, and are marked with corresponding timestamps; Determine the actual number of P-type target temporal hypergraph motifs pre-selected in the temporal cooperation hypergraph by using a counting algorithm, P ≥ 1; Creating a zero model random hypergraph set, and counting the average number of each type of target time series hypergraph motifs in the zero model random hypergraph set; Calculate the abundance of each type of target temporal hypergraph motif according to its true number in the temporal cooperation hypergraph and its average number in the zero model random hypergraph set; Based on the abundance of all target temporal hypergraph motifs, a characteristic profile of the temporal cooperation hypergraph is constructed, and the changing trend of the characteristic profile is analyzed to predict the evolution mechanism of the target industry.

2. According to claim 1, a method for predicting future industrial evolution mechanisms based on a temporal hypergraph motif is characterized in that: The original collaborative datasets between several institutions in the target industry include: Determine the target industry and extract the field keywords corresponding to the target industry; Using the keywords in the field as search conditions, obtain a list of institutions related to the target industry; Based on the list of institutions, cooperation events between institutions are collected to form an original cooperation data set.

3. According to claim 2, a method for predicting future industrial evolution mechanisms based on a temporal hypergraph motif is characterized in that: The preprocessing of the original cooperation data set to generate a cooperation event data set containing a timestamp includes: Disambiguate the names of institutions to ensure that the same institution has a unified name in all cooperation events; Assign a corresponding timestamp to each cooperation event according to the preset time unit; When multiple cooperation events have the same timestamp, unique timestamps are respectively configured for the multiple cooperation events in a random order.

4. According to claim 1, a method for predicting future industrial evolution mechanisms based on a temporal hypergraph motif is characterized in that: When the counting algorithm adopts an accurate counting algorithm based on temporal hypergraph projection, the counting algorithm is used to determine the true number of P-type target temporal hypergraph motifs pre-selected in the temporal cooperation hypergraph, P≥1, including: Given a time window δ, and initialize the projection map , where node v∈ The set of hyperedges with the same node set in the corresponding time-series cooperation hypergraph , the timestamp set carried by each node v = ,side ∈ If and only if ∩ ≠ ; Update based on inserting modules and deleting modules : In the insertion module, for a given timing hyperedge , if there is no hyperedge with the same node set in the current time window, then The corresponding node is added to and find the adjacent nodes of the newly added node To create an associated edge, set the timestamp Add to middle; In the deletion module, if a given timing hyperedge Is the only node set in the current time window The timing superedge of Remove the node and all its associated edges, and delete middle The corresponding timestamp; In the specified time window In the Count the ternary, binary, and unary temporal hypergraph motifs.

5. According to claim 4, a method for predicting future industrial evolution mechanisms based on a temporal hypergraph motif is characterized in that: When the counting algorithm adopts an approximate counting method based on random sampling, the counting algorithm is used to determine the true number of P-type target temporal hypergraph motifs pre-selected in the temporal cooperative hypergraph, P≥1, including: Set the sampling time window length ∆ and the number of sampling times s; For each sampling, the following steps are performed: a. From the temporal hyperedge set Randomly and uniformly select a temporal hyperedge ; b. According to the timestamp t of the selected temporal hyperedge, in the time window [ ]A random timestamp is selected ; c. Using the counting algorithm described in claim 4 to count the sub-windows [ ] to accurately count the temporal hypergraph motifs within ; d. Calculate the probability that a temporal hyperedge appears in the selected time window ,in, is the number of hyperedges in the subwindow; e. Adjust the counting result to ; The estimated value for each sample is summed and multiplied by , an unbiased estimate of each type of temporal hypergraph motif is obtained, and the unbiased estimate is taken as its true number.

6. A method for predicting future industrial evolution mechanisms based on a temporal hypergraph motif according to any one of claims 4 or 5, characterized in that: The step of creating a zero model random hypergraph set and counting the average number of each type of target time series hypergraph motifs in the zero model random hypergraph set includes: constructing a set of zero-model random hypergraphs using a zero-model generation method, wherein each random hypergraph retains basic statistical properties of the temporal cooperation hypergraph; According to the counting method described in any one of claims 4 or 5, the number of occurrences of each type of target time-series hypergraph motif in each random hypergraph is calculated, and the average value is recorded.

7. According to claim 1, a method for predicting future industrial evolution mechanism based on time-series hypergraph motifs is characterized in that: The characteristic profile of the temporal cooperation hypergraph is constructed based on the abundance of all target temporal hypergraph motifs, and the changing trend of the characteristic profile is analyzed to predict the evolution mechanism of the target industry, including: Calculate the normalized abundance of each type of target temporal hypergraph motif, and use the normalized abundance as a dimension to construct a feature profile of the temporal cooperation hypergraph; Analyze the differences in the characteristic profiles in different periods and predict the evolution mechanism of the target industry.

8. A device for predicting future industrial evolution mechanisms based on a temporal hypergraph motif, characterized in that: include: A preprocessing module is used to collect the original cooperation data set between several institutions in the target industry field, and preprocess the original cooperation data set to generate a cooperation event data set containing a timestamp; A hypergraph construction module, used to construct a time-series cooperation hypergraph based on the cooperation event dataset, wherein nodes represent organizations, hyperedges represent cooperation events, and are marked with corresponding timestamps; A motif detection module, used for determining the actual number of P-type target temporal hypergraph motifs pre-selected in the temporal cooperation hypergraph by using a counting algorithm, where P≥1; A statistical module, used to create a zero model random hypergraph set, and count the average number of each type of target time series hypergraph motifs in the zero model random hypergraph set; An abundance calculation module is used to calculate the abundance of each type of target time-series hypergraph motif according to its actual number in the time-series cooperation hypergraph and its average number in the zero model random hypergraph set; A prediction module is used to construct a characteristic profile of the temporal cooperation hypergraph based on the abundance of all target temporal hypergraph motifs, and analyze the changing trend of the characteristic profile to predict the evolution mechanism of the target industry.

9. An electronic device, characterized in that: It includes a memory and a processor, the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement a future industry evolution mechanism prediction method based on a time-series hypergraph motif as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: The computer program enables the computer to implement a method for predicting future industrial evolution mechanisms based on a time-series hypergraph motif as described in any one of claims 1 to 7 when executed.

Images