A method and device for calculating information correlation degree based on natural language processing

Abstract

The application discloses a kind of information correlation degree calculation method and device based on natural language processing, method includes: according to the technical content of service and internal technology in service domain, the first associated information between the service domain and the internal technology is constructed;According to the technical content of external technology and the technical content of the internal technology of self-driving traffic system, the second associated information between the external technology and the internal technology is constructed;After the service domain and the external technology are expressed as the vector about internal technology, the third associated information between the service domain and the external technology is constructed according to corresponding vector.The application can establish the correlation of service domain, internal technology and external technology of self-driving traffic system, can obtain internal technology and external technology closely associated with self-driving traffic system, and can be widely applied in traffic system data processing technical field.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of traffic system data processing, and in particular to an information correlation degree calculation method and device based on natural language processing. BACKGROUND

[0002] Transportation is a basic, leading and service industry of the national economy, and plays an important role in optimizing the national industrial layout, reducing development costs and reducing environmental pollution. Autonomous transportation system (ATS) is based on the update and development of the existing intelligent transportation system framework in China, aiming to reduce human intervention and realize the self-organization operation and autonomous service of the transportation system. The biggest difference between ATS and traditional intelligent transportation system is the high autonomy of the system. People, vehicles, roads and environment are the four basic elements of the transportation system. ATS realizes the integration of people, vehicles, roads and environment through the operation process of autonomous perception, autonomous learning, autonomous decision-making and autonomous response, and guarantees the orderly operation of road traffic.

[0003] The basic theoretical method of autonomous transportation system can be used for the establishment of comprehensive transportation system architecture. Among them, the construction of the correlation relationship of the elements of the autonomous transportation system helps to study the influence of element evolution on each other and the influence of element evolution on the transportation system, but there is no perfect correlation degree calculation method for the elements of the autonomous transportation system. SUMMARY

[0004] Therefore, the embodiments of the present application provide an information correlation degree calculation method and device based on natural language processing.

[0005] One aspect of the present application provides an information correlation degree calculation method based on natural language processing, comprising:

[0006] According to the technical content of the services and internal technologies in the service domain, first correlation information between the service domain and the internal technologies is constructed;

[0007] According to the technical content of the external technologies and the internal technologies of the autonomous transportation system, second correlation information between the external technologies and the internal technologies is constructed;

[0008] After the service domain and the external technologies are represented as vectors about internal technologies, third correlation information between the service domain and the external technologies is constructed according to the corresponding vectors.

[0009] Optionally, the first correlation information between the service domain and the internal technologies is constructed according to the technical content of the services and internal technologies in the service domain, comprising:

[0010] construct a retrieval formula according to services in a service domain, the retrieval formula being used to retrieve patent data;

[0011] extract internal technologies from the retrieved patent data using a latent Dirichlet topic model;

[0012] calculate a degree of association between the service domain and the internal technologies according to a belonging relationship between the service domain and the patents and a probability relationship between the internal technologies and the patents, as the first association information;

[0013] classify the internal technologies according to the first association information.

[0014] Optionally, the extracting the internal technologies from the retrieved patent data using the latent Dirichlet topic model comprises: extracting the internal technologies from the patent data using the latent Dirichlet topic model and obtaining the probability relationship between the internal technologies and the patents;

[0015] The calculating the degree of association between the service domain and the internal technologies according to the belonging relationship between the service domain and the patents and the probability relationship between the internal technologies and the patents comprises: calculating a probability that each internal technology belongs to each service domain in combination with the probability relationship between the internal technologies and the patents, taking the probability as an association degree value between the internal technology and the service domain, and the association degree value being calculated according to a formula:

[0016]

[0017] wherein, pro(s, k) represents the probability that the internal technology k belongs to the service domain s; M s represents a number of patents of the service domain s; p(k / d i (s) represents the probability that the internal technology k belongs to an i-th patent in the service domain s;

[0018] The classifying the internal technologies according to the first association information comprises: for each internal technology, taking a service domain with the largest probability as a classification of the internal technology.

[0019] Optionally, the constructing the second association information between the external technologies of the autonomous traffic system and the internal technologies according to the technical content of the external technologies of the autonomous traffic system and the technical content of the internal technologies comprises:

[0020] defining the external technologies of the autonomous traffic system;

[0021] performing text vectorization processing on the external technologies and the patents to obtain vector information of the external technologies and the patents;

[0022] calculating a text similarity between the external technologies and the patents according to the vector information;

[0023] According to the text similarity of the external technology and the patent and the probability relationship of the internal technology and the patent, a correlation degree of the internal technology and the external technology is calculated as the second correlation information.

[0024] Optionally, the external technology of the autonomous traffic system comprises: adopting a top-down classification idea to find the external technology of the autonomous traffic system from five aspects of autonomous perception, autonomous learning, autonomous decision, autonomous response and auxiliary operation, establishing an external technology list, and giving a definition paradigm of the external technology as the text of the external technology.

[0025] The text vectorization processing of the external technology and the patent to obtain vector information of the external technology and the patent comprises: performing text vectorization on the text of the patent and the text of the external technology based on a TF-IDF algorithm to obtain TF-IDF vectors of the patent and the external technology.

[0026] The calculation formula of the TF-IDF vector is:

[0027]

[0028] Wherein, N dj is the frequency of occurrence of word j in document d, N d is the total number of words in document d, M is the total number of documents, and M j is the number of documents in which word j appears.

[0029] The text similarity of the external technology and the patent is calculated according to the vector information, comprising: calculating the cosine similarity of the TF-IDF vector of each patent and each external technology.

[0030] The calculation formula of the cosine similarity is:

[0031]

[0032] Wherein, sim(d, e) is the text similarity of the patent d and the external technology e; and cos(d_text, e_text) is the cosine similarity of the text vector d_text and the text vector e_text.

[0033] According to the text similarity of the external technology and the patent and the probability relationship of the internal technology and the patent, the correlation degree of the internal technology and the external technology is calculated, comprising: according to the probability relationship of the internal technology and the patent and the text similarity of the patent and the external technology, the correlation degree of each internal technology and each external technology is calculated.

[0034] The calculation formula of the correlation degree of each internal technology and each external technology is:

[0035]

[0036] wherein cor(k, e) is the correlation degree of the internal technology k and the external technology e; p(k / d i ) is the probability that the internal technology k belongs to the i-th patent; sim(d i , e) is the text similarity of the i-th patent and the external technology e; and M is the total number of patents.

[0037] Optionally, after the service domain and the external technology are represented as vectors with respect to the internal technology, third correlation information between the service domain and the external technology is constructed according to the corresponding vectors, including:

[0038] representing the service domain and the external technology as vectors with respect to the internal technology;

[0039] performing data standardization processing according to the vectors to obtain a service domain vector and an external technology vector;

[0040] calculating the Euclidean distance between the service domain vector and the external technology vector according to the result of the standardization processing;

[0041] calculating the correlation degree of the service domain and the external technology according to the Euclidean distance.

[0042] Optionally, in the step of performing data standardization processing according to the vectors to obtain a service domain vector and an external technology vector, the standardization processing is 0-mean standardization, and the calculation formula of the 0-mean standardization is:

[0043]

[0044] wherein x norm is the result of the 0-mean standardization; μ is the mean, and σ is the standard deviation;

[0045] the calculation formula of the Euclidean distance is:

[0046]

[0047] wherein Euc(s_intec norm , e_intec norm ) is the Euclidean distance of the standardized vectors of the service domain s and the external technology e; s_intec norm is the standardized vector of the service domain s with respect to the internal technology; and e_intec norm is the standardized vector of the external technology e with respect to the internal technology;

[0048] the calculation formula of the correlation degree of the service domain and the external technology is:

[0049]

[0050] Wherein, C(s, e) is the correlation degree of service domain s and external technology e, the greater C(s, e) is, the stronger the correlation between service domain s and external technology e is; max(Euc) is the maximum value of all Euclidean distances; min(Euc) is the minimum value of all Euclidean distances.

[0051] The embodiment of the present application also provides another aspect of a natural language processing based information correlation degree calculation device, which comprises:

[0052] A first module is configured to construct first correlation information between a service domain and an internal technology according to technical content of services in the service domain and the internal technology;

[0053] A second module is configured to construct second correlation information between an external technology of an autonomous traffic system and the internal technology according to technical content of the external technology and the internal technology;

[0054] A third module is configured to construct third correlation information between the service domain and the external technology according to corresponding vectors after representing the service domain and the external technology as vectors about the internal technology.

[0055] The embodiment of the present application also provides another aspect of an electronic device, which comprises a processor and a memory;

[0056] The memory is configured to store a program;

[0057] The processor executes the program to realize the method as described above.

[0058] The embodiment of the present application also provides another aspect of a computer readable storage medium, which stores a program, and the program is executed by a processor to realize the method as described above.

[0059] The embodiment of the present application constructs first correlation information between a service domain and an internal technology according to technical content of services in the service domain and the internal technology, constructs second correlation information between an external technology of an autonomous traffic system and the internal technology according to technical content of the external technology and the internal technology, and constructs third correlation information between the service domain and the external technology according to corresponding vectors after representing the service domain and the external technology as vectors about the internal technology. The present application can establish the correlation relationship between the service domain, the internal technology and the external technology of the autonomous traffic system, and can obtain the internal technology and the external technology closely related to the autonomous traffic system. BRIEF DESCRIPTION OF DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort based on these drawings.

[0061] Figure 1 The overall flowchart provided for the embodiments of the present application;

[0062] Figure 2 The visualization result schematic diagram of the service domain-internal technology association provided for the embodiments of the present application;

[0063] Figure 3 The visualization result schematic diagram of the internal technology-external technology association provided for the embodiments of the present application;

[0064] Figure 4 The visualization result schematic diagram of the service domain-external technology association provided for the embodiments of the present application. DETAILED DESCRIPTION

[0065] In order to make the purposes, technical solutions and advantages of the present application more clear, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0066] In order to solve the technical problems existing in the prior art, the embodiments of the present application provide an information correlation degree calculation method based on natural language processing, comprising:

[0067] According to the technical content of the services and internal technologies in the service domain, the first association information between the service domain and the internal technologies is constructed;

[0068] According to the technical content of the external technologies and the internal technologies of the autonomous traffic system, the second association information between the external technologies and the internal technologies is constructed;

[0069] After the service domain and the external technologies are represented as vectors about internal technologies, the third association information between the service domain and the external technologies is constructed according to the corresponding vectors.

[0070] Optionally, the first association information between the service domain and the internal technologies is constructed according to the technical content of the services and internal technologies in the service domain, comprising:

[0071] The retrieval formula is constructed according to the services in the service domain, and the retrieval formula is used to retrieve patent data;

[0072] extracting internal technologies from the retrieved patent data using a latent Dirichlet allocation topic model;

[0073] According to the belonging relationship of the service domain and the patent and the probability relationship of the internal technology and the patent, a correlation degree of the service domain and the internal technology is calculated as the first correlation information;

[0074] According to the first correlation information, the internal technologies are classified.

[0075] Optionally, the extracting internal technologies from the retrieved patent data using a latent Dirichlet allocation topic model comprises: extracting internal technologies from the patent data using a latent Dirichlet allocation topic model, and obtaining a probability relationship of the internal technologies and the patents;

[0076] The calculating the correlation degree of the service domain and the internal technology according to the belonging relationship of the service domain and the patent and the probability relationship of the internal technology and the patent comprises: combining the probability relationship of the internal technologies and the patents, calculating a probability of each internal technology belonging to each service domain, and taking the probability as a correlation degree value of the internal technology and the service domain, and a calculation formula of the correlation degree value is:

[0077]

[0078] wherein, pro(s, k) represents the probability of the internal technology k belonging to the service domain s; M s represents the number of patents of the service domain s; p(k / d i (s) represents the probability of the internal technology k belonging to the i th patent in the service domain s;

[0079] The classifying the internal technologies according to the first correlation information comprises: for each internal technology, taking the service domain with the maximum probability as the classification of the internal technology.

[0080] Optionally, the constructing the second correlation information between the external technologies of the autonomous traffic system and the internal technologies according to the technical content of the external technologies and the technical content of the internal technologies comprises:

[0081] defining external technologies of the autonomous traffic system;

[0082] performing text vectorization processing on the external technologies and the patents to obtain vector information of the external technologies and the patents;

[0083] calculating a text similarity of the external technologies and the patents according to the vector information;

[0084] According to the text similarity of the external technologies and the patents and the probability relationship of the internal technologies and the patents, a correlation degree of the internal technologies and the external technologies is calculated as the second correlation information.

[0085] Optionally, the definition of the external technology of the autonomous traffic system comprises: adopting a top-down classification idea to find the external technology of the autonomous traffic system from five aspects of autonomous perception, autonomous learning, autonomous decision, autonomous response and auxiliary operation, establishing a list of external technologies, and giving a definition paradigm of the external technology as a text of the external technology;

[0086] The text vectorization processing of the external technology and the patent to obtain vector information of the external technology and the patent comprises: text vectorization of the text of the patent and the text of the external technology based on a TF-IDF algorithm to obtain a TF-IDF vector of the patent and the external technology:

[0087] The calculation formula of the TF-IDF vector is:

[0088]

[0089] Wherein, N dj is the frequency of occurrence of word j in document d, N d is the total number of words in document d, M is the total number of documents, M j is the number of documents in which word j appears;

[0090] The calculation of the text similarity between the external technology and the patent according to the vector information comprises: calculating the cosine similarity of the TF-IDF vector of each patent and each external technology;

[0091] The calculation formula of the cosine similarity is:

[0092]

[0093] Wherein, sim(d, e) is the text similarity between patent d and external technology e; cos(d_text, e_text) is the cosine similarity of the text vector d_text and the text vector e_text;

[0094] The calculation of the correlation degree between the internal technology and the external technology according to the text similarity between the external technology and the patent and the probability relationship between the internal technology and the patent comprises: calculating the correlation degree between each internal technology and each external technology according to the probability relationship between the internal technology and the patent and the text similarity between the patent and the external technology;

[0095] The calculation formula of the correlation degree between each internal technology and each external technology is:

[0096]

[0097] Wherein, cor(k, e) is the correlation degree between the internal technology k and the external technology e; p(k / d i) is the probability that the internal technology k belongs to the i-th patent; sim(d i , e) is the text similarity of the i-th patent and the external technology e; M is the total number of patents.

[0098] Optionally, after representing the service domain and the external technology as vectors with respect to the internal technology, the third association information between the service domain and the external technology is constructed according to the corresponding vectors, including:

[0099] representing the service domain and the external technology as vectors with respect to the internal technology;

[0100] performing data standardization processing according to the vectors to obtain a service domain vector and an external technology vector;

[0101] calculating the Euclidean distance between the service domain vector and the external technology vector according to the result of the standardization processing;

[0102] calculating the association degree between the service domain and the external technology according to the Euclidean distance.

[0103] Optionally, in the step of performing data standardization processing according to the vectors to obtain a service domain vector and an external technology vector, the standardization processing is 0-mean standardization, and the calculation formula of the 0-mean standardization is:

[0104]

[0105] wherein, x norm is the result of the 0-mean standardization; μ is the mean, and σ is the standard deviation;

[0106] the calculation formula of the Euclidean distance is:

[0107]

[0108] wherein, Euc(s_intec norm , e_intec norm ) is the Euclidean distance of the standardized vectors of the service domain s and the external technology e; s_intec norm is the standardized vector of the service domain s with respect to the internal technology; e_intec norm is the standardized vector of the external technology e with respect to the internal technology;

[0109] the calculation formula of the association degree between the service domain and the external technology is:

[0110]

[0111] Where C(s,e) is the correlation between service domain s and external technology e. The larger C(s,e) is, the stronger the correlation between service domain s and external technology e. max(Euc) is the maximum value among all Euclidean distances. min(Euc) is the minimum value among all Euclidean distances.

[0112] Another embodiment of the present invention provides an information relevance calculation device based on natural language processing, comprising:

[0113] The first module is used to construct first association information between the service domain and the internal technology based on the technical content of the services and internal technologies in the service domain;

[0114] The second module is used to construct a second association information between the external technology and the internal technology of the autonomous transportation system based on the technical content of the external technology and the technical content of the internal technology.

[0115] The third module is used to represent the service domain and the external technology as vectors about the internal technology, and then construct a third association information between the service domain and the external technology based on the corresponding vectors.

[0116] Another embodiment of the present invention provides an electronic device, including a processor and a memory;

[0117] The memory is used to store programs;

[0118] The processor executes the program to implement the method described above.

[0119] Another aspect of this invention provides a computer-readable storage medium storing a program that is executed by a processor to implement the method described above.

[0120] The specific implementation process of the present invention will now be described in detail with reference to the accompanying drawings:

[0121] like Figure 1 As shown, Figure 1 A method for calculating the correlation between service domains, internal technologies, and external technologies in an autonomous transportation system based on natural language processing includes the following steps:

[0122] Step S1, Service Domain-Internal Technology Association: Construct a retrieval query based on the services within the service domain to search for patent data, and extract internal technologies from the patent data using a Hidden Dirichlet Topic Model; calculate the degree of association between the service domain and internal technologies based on the attribution relationship between the service domain and the patents, as well as the probabilistic relationship between internal technologies and patents; further classify the internal technologies. The service domain-internal technology association results are as follows: Figure 2 As shown, Figure 2The middle node represents 12 service domains and 85 internal technologies. The color of the internal technology node represents the classification into the corresponding service domain. The color of the connection between the service domain and the internal technology represents the size of the degree of association. The figure filters the edges with a degree of association less than 0.02.

[0123] The service domain-internal technology association step in step S1 is as follows:

[0124] Step 101, extract internal technologies: search patent data according to the service in the service domain of the transportation system, and extract internal technologies from the patent data by using the latent Dirichlet topic model. The probability relationship between the internal technology and the patent can be obtained.

[0125] Specifically, first, the transportation system patent data is collected. ARC-IT 9.0 (Architecture Reference for Cooperative and Intelligent Transportation, ARC-IT) provides a general architecture for planning, defining, and integrating transportation systems, and provides 12 service domains. The present application uses the scope of the transportation system defined by ARC-IT 9.0 to obtain the internal technology of the transportation system. Based on the 150 service packages provided by ARC-IT 9.0, the present application constructs 150 search expressions, and retrieves 40288 patents with application dates from 2000 to 2019 from the relevant database.

[0126] Further, the present application uses the latent Dirichlet topic model to extract 85 topics from the titles and abstracts of the 40288 patents. Since the patents contain rich technical information, the 85 extracted topics are regarded as 85 internal technologies. The output of the latent Dirichlet topic model includes the probability distribution of document-topic and topic-word. Therefore, the internal technology can be named according to the high-probability words and high-probability document content.

[0127] Step 102, degree of association calculation: since the patent is obtained by constructing a search expression from the service in the service domain, the ownership relationship between the patent and the service domain is known. Combined with the probability relationship between the internal technology and the patent in step 101, the probability of each internal technology belonging to each service domain is calculated, and the probability is taken as the association degree value of the internal technology and the service domain.

[0128]

[0129] wherein pro(s, k) represents the probability of internal technology k belonging to service domain s, M s represents the number of patents of service domain s, p(k / d i (s) represents the probability of internal technology k belonging to the i-th patent in service domain s.

[0130] Step 103, technology classification: for each internal technology, the service domain with the highest probability is taken as the classification of the internal technology.

[0131] Step S2, internal technology-external technology association: define the external technology of the autonomous transportation system; text vectorize the external technology and the patent; calculate the text similarity between the external technology and the patent; according to the text similarity between the external technology and the patent and the probability relationship between the internal technology and the patent, calculate the association degree between the internal technology and the external technology. The internal technology-secondary external technology association result is as shown in Figure 3 Figure 3 In the figure, the nodes of the outer ring represent 43 external technologies, the nodes of the central radial structure represent 85 internal technologies, and the color depth of the connection between the internal technology and the external technology represents the size of the association degree. The figure filters the edges with an association degree less than 11.

[0132] The internal technology-secondary external technology association step in step S2 is:

[0133] Step 201, define external technology: adopt a top-down classification idea, find ATS external technology from five aspects of autonomous perception, autonomous learning, autonomous decision-making, autonomous response, and auxiliary operation, establish a list of external technologies, and give the definition paradigm of secondary external technology as the text of external technology.

[0134] Specifically, the autonomous perception link collects information data of the entity world user and the environment, which is the basis of ATS. Facing the entity world composed of people, vehicles, roads, and environment, ATS obtains information in the autonomous perception link as data input for the autonomous learning link. The first-level external technology of the autonomous perception link is divided into positioning technology and sensor technology. Positioning technology helps users obtain their exact location, and sensor technology helps users obtain environmental information around them.

[0135] The autonomous learning link extracts effective information and knowledge based on data. According to the entity world data collected by the autonomous perception link, ATS extracts information in the autonomous learning link to enable the system to accurately understand. The first-level external technology of the autonomous learning link includes big data technology and computing technology. Big data technology can help users obtain information from a large variety of data, and computing technology can help users quickly and accurately obtain information.

[0136] The autonomous decision-making link forms decisions based on information and knowledge. Based on the information and knowledge processed by the autonomous learning link, in the autonomous decision-making link, ATS accurately predicts the possible situations that may occur and accurately judges and plans the next action of the system, making reasonable decisions on the behavior of the system. The first-level external technology of the autonomous decision-making link includes artificial intelligence technology, which can help the system simulate human thinking to make decisions.​

[0137] The autonomous response link forms control over the real world based on decisions. In the autonomous response link, the system implements the decisions and plans made by the autonomous decision link into actual behaviors, achieving the purpose of autonomous traffic system. The first-level external technology of the autonomous response link includes control technology, which can control the behavior of users.

[0138] The effective operation of ATS cannot be separated from the functions of the four links of autonomous perception, autonomous learning, autonomous decision, and autonomous response. The internal technology of the four links is needed to support the operation of the links. In addition, information transmission is needed between the four links, and the entire system needs to ensure effective operation, so the technology that assists the operation of the system is needed. The first-level external technology of the auxiliary operation link includes communication technology and digital simulation technology. The communication technology can realize the data and information transmission of the system, and the digital simulation technology can simulate the operation status of the system.

[0139] Finally, the present application divides the external technology into 8 first-level external technologies, and further obtains 43 second-level external technologies on this basis. In addition, the concept of the second-level external technology is given in a certain format of paradigm, as the text of the second-level external technology.

[0140] Step 202, text vectorization: based on the TF-IDF algorithm (Term Frequency-Inverse Document Frequency, TF-IDF), the text of the patent and the text of the second-level external technology are text vectorized to obtain the TF-IDF vector of the patent and the second-level external technology.

[0141]

[0142] wherein, N dj is the frequency of occurrence of word j in document d, N d is the total number of words in document d, M is the total number of documents, and M j is the number of documents in which word j appears.

[0143] Step 203, cosine similarity calculation: the cosine similarity of the TF-IDF vector of each patent and each second-level external technology is calculated.

[0144]

[0145] wherein, sim(d, e) is the text similarity of patent d and second-level external technology e, and cos(d_text, e_text) is the cosine similarity of the text vector d_text and the text vector e_text.

[0146] Step 204, correlation calculation: according to the probability relationship between the internal technology and the patent obtained in step 101 and the text similarity between the patent and the secondary external technology obtained in step 203, the correlation between each internal technology and each secondary external technology is calculated.

[0147]

[0148] wherein cor(k, e) is the correlation between the internal technology k and the secondary external technology e, p(k / d i ) is the probability that the internal technology k belongs to the i-th patent, sim(d i , e) is the text similarity between the i-th patent and the secondary external technology e, and M is the total number of patents.

[0149] Step S3, service domain-external technology correlation: the service domain and the external technology are both expressed as vectors about the internal technology, and data standardization processing is performed; the Euclidean distance between the service domain vector and the external technology vector is calculated; and the correlation between the service domain and the secondary external technology is calculated. The service domain-external technology correlation result is shown in Figure 4 , wherein the color depth represents the correlation degree between the service domain and the external technology. Figure 4

[0150] The service domain-secondary external technology correlation step in step S3 is as follows:

[0151] Step 301, data standardization: according to the correlation between the service domain and the internal technology obtained in step 102 and the correlation between the internal technology and the secondary external technology obtained in step 204, the service domain and the secondary external technology are both expressed as vectors about the internal technology, and the service domain vector and the secondary external technology vector are 0-mean standardized for subsequent calculation.

[0152]

[0153] wherein μ is the mean and σ is the standard deviation.

[0154] Step 302, Euclidean distance calculation: the Euclidean distance between the 0-mean standardized service domain vector and the secondary external technology vector is calculated.

[0155]

[0156] wherein Euc(s_intec norm , e_intec norm ) is the Euclidean distance of the standardized vectors of the service domain s and the secondary external technology e, s_intec norm is the standardized vector of the service domain s about the internal technology, and e_intec norm ​is a secondary external technology e about the internal technology standardization vector.

[0157] Step 303, correlation degree calculation: according to step 302, the Euclidean distance of the service domain vector and the secondary external technology vector is calculated to calculate the correlation degree of the service domain and the secondary external technology.

[0158]

[0159] Wherein, C(s, e) is the correlation degree of the service domain s and the external technology e, the greater C(s, e) represents the stronger correlation between the service domain s and the external technology e, max(Euc) is the maximum value of all Euclidean distances, and min(Euc) is the minimum value of all Euclidean distances.

[0160] In summary, the first correlation information between the service domain and the internal technology is constructed according to the technical content of the service in the service domain and the internal technology, the second correlation information between the external technology and the internal technology is constructed according to the technical content of the external technology of the autonomous traffic system and the technical content of the internal technology, and the third correlation information between the service domain and the external technology is constructed according to the corresponding vectors after the service domain and the external technology are represented as vectors about the internal technology. The correlation relationship between the service domain, the internal technology and the external technology of the autonomous traffic system can be established, and the internal technology and the external technology closely related to the autonomous traffic system can be obtained.

[0161] In some alternative embodiments, the functions / operations mentioned in the block diagram can not occur in the order mentioned in the operation diagram. For example, depending on the functions / operations involved, two blocks shown in succession can actually be executed substantially simultaneously or the blocks can sometimes be executed in reverse order. In addition, the embodiments presented and described in the flowcharts of the present application are provided by way of example, and the purpose is to provide a more comprehensive understanding of the technology. The disclosed method is not limited to the operations and logical flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and in which sub-operations described as part of larger operations are independently executed.

[0162] Furthermore, although the present application is described in the context of functional modules, it is to be understood that one or more of the described functions and / or features can be integrated in a single physical device and / or software module, or one or more functions and / or features can be implemented in separate physical devices or software modules. It will also be appreciated that detailed discussion of the actual implementation of each module is not necessary to an understanding of the application. Rather, the actual implementation of the modules, in combination with their attributes, functions, and internal relationships, are to be understood within the context of the devices disclosed herein. Thus, those skilled in the art with access to patents, scientific journals, and other public sources known by those skilled in the art will be able to practice the application as set forth in the claims without undue experimentation, using ordinary skill in the art along with the present disclosure. It is also to be understood that the specific concepts disclosed are merely illustrative and that the scope of the present application is to be determined by the entire scope of the claims, along with all equivalents of the claims and their equivalents.

[0163] If the functions are implemented in software, the functions can be stored in or implemented as one or more computer program products, which can be incorporated into a computer-readable medium for use by or in connection with an apparatus, method or system as described herein. The computer-readable medium can be a machine-readable storage device, a machine-readable storage substrate, a memory device, or a computer-readable signal. The computer-readable medium can be, for example, but is not limited to, volatile or non-volatile memory, a floppy diskette, a compact disc read-only memory (CD-ROM), a magnetic tape, a flash drive, a hard disk drive, a programmable read-only memory (PROM), a random access memory (RAM), a reprogrammable read-only memory (REPROM) or a similar medium. The computer-readable medium can be paper, on which the program is printed. The computer-readable medium can be a medium, such as RAM, ROM, PROM, a tape, a hard disk or a compact disc, which can store data for use by or in connection with an application program. The computer-readable medium can also be a medium that carries such data also, or a medium that merges, for use by or in connection with an application program. The computer-readable medium can also be paper or any other medium, on which the program is printed.

[0164] The logic and / or steps represented in the flowcharts and / or otherwise described herein, for example, can be considered as a sequence of executable instructions stored in a computer readable medium, which can be executed by an instruction execution system, apparatus or device, such as a computer-based system, a processor-based system, or other system that can fetch the instructions from the instruction execution system, apparatus or device and execute the instructions. For the purposes of this specification, a "computer readable medium" can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus or device.

[0165] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection (electronic) having one or more wires, a portable computer diskette (magnetic), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can also be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example, via optical scanning of the paper or other medium, then compiled, interpreted, or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

[0166] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the embodiments described above, various steps or methods can be implemented, for example, by software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and in another embodiment, any of the following techniques can be used to implement the hardware used in the described embodiments: discrete logic circuitry having logic gates for implementing logic functions upon data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), and so forth.

[0167] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" and the like means that the specific feature, structure, material or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In the specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any suitable manner in one or more embodiments or examples.

[0168] Although embodiments of the application have been shown and described, it would be recognized by those of ordinary skill in the art that various changes, modifications, substitutions, and alterations can be made hereto without departing from the spirit and scope of the application. The scope of the application is limited only by the claims and the equivalents thereof.

[0169] The above is a specific description of the preferred embodiments of the present application, but the present application is not limited to the described embodiments, and those skilled in the art can make various equivalent modifications or replacements without departing from the spirit of the present application, and these equivalent modifications or replacements are all included in the scope defined by the claims of the present application.

Claims

1. A method for calculating information relevance based on natural language processing, characterized in that, include: Based on the technical content of the services and internal technologies in the service domain, construct the first association information between the service domain and the internal technologies; Based on the technical content of the external technology and the technical content of the internal technology of the autonomous transportation system, a second association information between the external technology and the internal technology is constructed; After representing the service domain and the external technology as vectors about the internal technology, a third association information between the service domain and the external technology is constructed based on the corresponding vectors. The step of constructing the first association information between the service domain and the internal technology based on the technical content of the services and internal technologies in the service domain includes: A search query is constructed based on services in the service domain, and the search query is used to retrieve patent data; The hidden Dirichlet theme model was used to extract the internal technology from the retrieved patent data; Based on the attribution relationship between service domains and patents, and the probabilistic relationship between internal technologies and patents, the degree of association between service domains and internal technologies is calculated as the first association information; The internal technologies are classified according to the first association information; The construction of a second association information between the external and internal technologies based on the technical content of the external technologies and the technical content of the internal technologies of the autonomous transportation system includes: External technologies that define autonomous transportation systems; External technologies and patents are vectorized into text to obtain vector information of external technologies and patents; The textual similarity between the external technology and the patent is calculated based on the vector information. Based on the textual similarity between external technologies and patents, and the probabilistic relationship between internal technologies and patents, the degree of association between internal technologies and external technologies is calculated as the second association information; After representing the service domain and the external technology as vectors about the internal technology, the third association information between the service domain and the external technology is constructed based on the corresponding vectors, including: Both service domains and external technologies are represented as vectors about internal technologies; Data standardization is performed on the vectors to obtain service domain vectors and external technology vectors; Based on the results of the standardization process, calculate the Euclidean distance between the service domain vector and the external technology vector; The degree of correlation between the service domain and external technologies is calculated based on the Euclidean distance.

2. The method for calculating information relevance based on natural language processing according to claim 1, characterized in that, The step of using a Hidden Dirichlet Theme Model to extract internal technologies from the retrieved patent data includes: using a Hidden Dirichlet Theme Model to extract internal technologies from the patent data and obtaining the probabilistic relationship between the internal technologies and the patents. The step of calculating the correlation between a service domain and internal technologies based on the attribution relationship between service domains and patents, and the probabilistic relationship between internal technologies and patents, includes: combining the probabilistic relationship between internal technologies and patents to calculate the probability that each internal technology belongs to each service domain, and using this probability as the correlation value between the internal technology and the service domain. The formula for calculating the correlation value is as follows: Where pro(s,k) represents the probability that internal technology k belongs to service domain s; M s The number of patents representing service domain s; p(k / d i (s) ) represents the probability that internal technology k belongs to the i-th patent in service domain s; The step of classifying the internal technologies based on the first association information includes: for each internal technology, taking the service domain with the highest probability as the classification of the internal technology.

3. The method for calculating information relevance based on natural language processing according to claim 1, characterized in that, The definition of external technologies for autonomous transportation systems includes: adopting a top-down classification approach to identify external technologies for autonomous transportation systems from five aspects: autonomous perception, autonomous learning, autonomous decision-making, autonomous response, and assisted operation; establishing a list of external technologies; and providing a definition paradigm for the external technologies as the text of the external technologies. The process of vectorizing external technologies and patents to obtain vector information of external technologies and patents includes: Based on the TF-IDF algorithm, the text of the patent and the text of the external technology are vectorized to obtain the TF-IDF vectors of the patent and the external technology; The formula for calculating the TF-IDF vector is as follows: Where, N dj N is the frequency of word j in document d. d M is the total number of words in document d, and M is the total number of documents. j It is the number of documents containing the word 'j'; The step of calculating the text similarity between the external technology and the patent based on the vector information includes: calculating the cosine similarity of the TF-IDF vectors of each patent and each external technology; The formula for calculating the cosine similarity is as follows: Where sim(d,e) is the text similarity between patent d and external technology e; cos(d_text,e_text) is the cosine similarity between text vector d_text and text vector e_text; The step of calculating the correlation between internal and external technologies based on the textual similarity between external technologies and patents and the probabilistic relationship between internal technologies and patents includes: calculating the correlation between each internal technology and each external technology based on the probabilistic relationship between internal technologies and patents and the textual similarity between patents and external technologies. The formula for calculating the correlation between each internal technology and each external technology is as follows: Where cor(k,e) is the correlation between internal technology k and external technology e; p(k / d i ) is the probability that internal technology k belongs to the i-th patent; sim(d i ,e) is the text similarity between the i-th patent and external technology e; M is the total number of patents.

4. The method for calculating information relevance based on natural language processing according to claim 1, characterized in that, In the step of performing data standardization processing based on the vector to obtain the service domain vector and the external technology vector, the standardization processing is zero-mean standardization, and the calculation formula for zero-mean standardization is: Where, x norm This is the result of zero-mean standardization; μ is the mean, and σ is the standard deviation. The formula for calculating the Euclidean distance is: Among them, Euc(s_intec norm e_intec norm ) is the Euclidean distance between the normalized vectors of the service domain s and the external technology e; s_intec norm It is the normalized vector of service domain s with respect to its internal technology; e_intec norm It is the standardized vector of external technology e with respect to internal technology; The formula for calculating the degree of correlation between the service domain and external technologies is as follows: Where C(s,e) is the correlation between service domain s and external technology e. The larger C(s,e) is, the stronger the correlation between service domain s and external technology e. max(Euc) is the maximum value among all Euclidean distances. min(Euc) is the minimum value among all Euclidean distances.

5. An information correlation calculation device based on natural language processing, characterized in that, include: The first module is used to construct first association information between the service domain and the internal technology based on the technical content of the services and internal technologies in the service domain; The second module is used to construct a second association information between the external technology and the internal technology of the autonomous transportation system based on the technical content of the external technology and the technical content of the internal technology. The third module is used to represent the service domain and the external technology as vectors about the internal technology, and then construct a third association information between the service domain and the external technology based on the corresponding vectors. Specifically, the first module is used for: A search query is constructed based on services in the service domain, and the search query is used to retrieve patent data; The hidden Dirichlet theme model was used to extract the internal technology from the retrieved patent data; Based on the attribution relationship between service domains and patents, and the probabilistic relationship between internal technologies and patents, the degree of association between service domains and internal technologies is calculated as the first association information; The internal technologies are classified according to the first association information; The second module is specifically used for: External technologies that define autonomous transportation systems; External technologies and patents are vectorized into text to obtain vector information of external technologies and patents; The textual similarity between the external technology and the patent is calculated based on the vector information. Based on the textual similarity between external technologies and patents, and the probabilistic relationship between internal technologies and patents, the degree of association between internal technologies and external technologies is calculated as the second association information; The third module is specifically used for: Both service domains and external technologies are represented as vectors about internal technologies; Data standardization is performed on the vectors to obtain service domain vectors and external technology vectors; Based on the results of the standardization process, calculate the Euclidean distance between the service domain vector and the external technology vector; The degree of correlation between the service domain and external technologies is calculated based on the Euclidean distance.

6. An electronic device, characterized in that, Including the processor and memory; The memory is used to store programs; The processor executes the program to implement the method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The storage medium stores a program that is executed by a processor to implement the method as described in any one of claims 1 to 4.

Images