Resource retrieval method and system based on big data

Abstract

The big data-based resource retrieval method and system provided in the present application refers to a first local innovation resource data cluster bound to a reference local innovation resource data cluster, and uses a pre-configured first artificial intelligence thread to process the reference local innovation resource data cluster and the first local innovation resource data cluster to obtain an optimized local innovation resource data cluster, and performs retrieval based on the local innovation resource data cluster to obtain a resource retrieval result. Thus, for the first artificial intelligence thread used to optimize the reference innovation resource data, only the configured innovation resource data obtained based on the reference innovation resource data is configured, thereby improving the problem of key feature similarity between different themes of the innovation resource data, reducing interference in the reference innovation resource data and reducing the probability of abnormalities in the reference innovation resource data labels. In this way, the accuracy and completeness of the resource retrieval results can be effectively improved.

Description

Technical Field

[0001] The present application relates to the field of big data and data retrieval technology, and more specifically, to a resource retrieval method and system based on big data. Background Art

[0002] Resource information retrieval is the process of people using specific search instructions, search terms and search strategies on a computer or a computer retrieval network terminal.

[0003] At present, the existing technology searches for resources through key features, but in actual application, the inventors found that some information to be searched has a high degree of similarity, and the results after the search are very similar. In this way, it is impossible to accurately obtain the resource search results, and it is difficult to ensure the accuracy and completeness of the resource search results. Summary of the invention

[0004] In view of this, the present application provides a resource retrieval method and system based on big data.

[0005] In a first aspect, a resource retrieval method based on big data is provided, which is applied to a resource retrieval system, and the method at least comprises:

[0006] Perform keyword mining on the reference innovation resource data to be retrieved to obtain local innovation resource data bound to the reference innovation resource data;

[0007] Determine a reference local innovative resource data cluster to be retrieved in the reference innovative resource data; determine a first local innovative resource data cluster bound to the reference local innovative resource data cluster in the local innovative resource data;

[0008] Processing the reference local innovative resource data cluster and the first local innovative resource data cluster using a pre-configured first artificial intelligence thread to obtain an optimized local innovative resource data cluster bound to the reference local innovative resource data cluster, wherein the first artificial intelligence thread is obtained by configuring the configured local innovative resource data cluster in advance, and the configured innovative resource data in the configured local innovative resource data cluster is obtained based on the reference innovative resource data;

[0009] A search is performed based on the local innovation resource data cluster to obtain a resource search result.

[0010] In an independently implemented embodiment, the configured local innovative resource data cluster includes a plurality of configured innovative resource data tuples, one of the plurality of configured innovative resource data tuples includes a first resource theme and a second resource theme, wherein the first resource theme is an innovative resource data theme screened from the reference innovative resource data, the second resource theme is an innovative resource data theme screened from the local innovative resource data, and the first resource theme and the second resource theme belonging to the consistently configured innovative resource data tuple correspond to consistently screened locations; before the reference local innovative resource data cluster and the first local innovative resource data cluster are processed by using a pre-configured first artificial intelligence thread, the method further includes:

[0011] For one of the several configured innovation resource data tuples, performing abnormal behavior query processing on the first resource topic and the second resource topic to obtain a first abnormal topic bound to the first resource topic and a second abnormal topic bound to the second resource topic;

[0012] Transmitting the first abnormal topic and the second abnormal topic to the first artificial intelligence thread, and obtaining a first optimized topic bound to the first abnormal topic based on the first artificial intelligence thread;

[0013] The first artificial intelligence thread is configured based on the similarity between the first optimization topic and the first resource topic.

[0014] In an independently implemented embodiment, performing abnormal behavior query processing on the first resource subject and the second resource subject to obtain a first abnormal subject bound to the first resource subject and a second abnormal subject bound to the second resource subject includes:

[0015] Generate an application scenario for binding the first resource theme and the second resource theme;

[0016] Performing application scenario identification processing on the first resource theme by using the application scenario to obtain a first abnormal theme bound to the first resource theme;

[0017] The second resource theme is subjected to application scenario identification processing by utilizing the application scenario to obtain a second abnormal theme bound to the second resource theme.

[0018] In an independently implemented embodiment, the transmitting the first abnormal topic and the second abnormal topic to the first artificial intelligence thread, and obtaining the first optimized topic bound to the first abnormal topic based on the first artificial intelligence thread, includes:

[0019] Transmitting the second abnormal topic to the first local thread of the first artificial intelligence thread, and obtaining a second optimized topic bound to the second abnormal topic based on the first local thread;

[0020] The first abnormal topic and the second optimized topic are transmitted to the second local thread of the first artificial intelligence thread, and the first optimized topic bound to the first abnormal topic is obtained based on the second local thread.

[0021] In an independently implemented embodiment, the first artificial intelligence thread configured by the similarity between the first optimization theme and the first resource theme includes: the first artificial intelligence thread configured by the similarity between the first optimization theme and the first resource theme, and the similarity between the second optimization theme and the second resource theme.

[0022] In an independently implemented embodiment, the processing of the reference local innovative resource data cluster and the first local innovative resource data cluster by using a pre-configured first artificial intelligence thread to obtain an optimized local innovative resource data cluster bound to the reference local innovative resource data cluster includes:

[0023] The first local innovative resource data cluster is transmitted to a first local thread of a pre-configured first artificial intelligence thread, and a second local innovative resource data cluster bound to the first local innovative resource data cluster is obtained based on the first local thread;

[0024] The reference local innovative resource data cluster and the second local innovative resource data cluster are processed by the second local thread of the first artificial intelligence thread to obtain an optimized local innovative resource data cluster bound to the reference local innovative resource data cluster.

[0025] In an independently implemented embodiment, the processing of the reference local innovative resource data cluster and the second local innovative resource data cluster by the second local thread of the first artificial intelligence thread to obtain an optimized local innovative resource data cluster bound to the reference local innovative resource data cluster includes:

[0026] Performing identification processing on the reference local innovative resource data cluster to obtain an identified reference local innovative resource data cluster;

[0027] The reference local innovative resource data cluster and the second local innovative resource data cluster after identification are transmitted to the second local thread of the first artificial intelligence thread, and the optimized local innovative resource data cluster bound to the reference local innovative resource data cluster is obtained based on the second local thread.

[0028] In an independently implemented embodiment, the identifying process of the reference local innovative resource data cluster to obtain the identified reference local innovative resource data cluster includes:

[0029] Determining a feature extraction set in the reference local innovative resource data cluster through the reference local innovative resource data cluster and the second local innovative resource data cluster;

[0030] determining reference features similar to features of the reference local innovation resource data cluster;

[0031] The reference features are extracted into the feature extraction set to obtain a reference local innovation resource data cluster after identification processing.

[0032] In an independently implemented embodiment, determining the feature extraction set in the reference local innovative resource data cluster through the reference local innovative resource data cluster and the second local innovative resource data cluster includes:

[0033] The constraints to be optimized in the reference local innovation resource data cluster and the significance identification conditions in the second local innovation resource data cluster are detected and processed to determine the reference optimization constraints in the reference local innovation resource data cluster, wherein the constraints to be optimized represent the constraints that are not optimized among the abnormal constraints in the reference local innovation resource data cluster; the constraints other than the reference optimization constraints in the constraints to be optimized are determined as feature extraction sets.

[0034] In an independently implemented embodiment, the constraints to be optimized in the reference local innovation resource data cluster and the significant identification conditions in the second local innovation resource data cluster are detected and processed to determine the reference optimization constraints in the reference local innovation resource data cluster, including: compressing the constraints to be optimized in the reference local innovation resource data cluster to obtain the first application scenario constraints; expanding the significant identification conditions in the second local innovation resource data cluster to obtain the second application scenario constraints; and determining the fusion result of the first application scenario constraints and the second application scenario constraints as the reference optimization constraints in the reference local innovation resource data cluster.

[0035] In an independently implemented embodiment, the determining of the reference local innovative resource data cluster to be retrieved from the reference innovative resource data includes: in response to the presence of no optimized constraints in the abnormal constraints of the reference innovative resource data, determining a reference distribution label in the no optimized constraints; and determining the reference local innovative resource data cluster to be retrieved from the reference innovative resource data through the reference distribution label, wherein the reference local innovative resource data cluster covers the reference distribution label.

[0036] In an independently implemented embodiment, determining a reference distribution label in the absence of optimization constraints includes: determining a significant random point in the absence of optimization constraints as a reference distribution label.

[0037] In an independently implemented embodiment, the reference distribution label is used to determine a reference local innovation resource data cluster to be retrieved from the reference innovation resource data, including: based on the reference distribution label, selecting a local innovation resource data cluster of a first previously set dimension from the reference innovation resource data as the reference local innovation resource data cluster to be retrieved.

[0038] In an independently implemented embodiment, after obtaining the optimized local innovative resource data cluster bound to the reference local innovative resource data cluster, the method further includes: based on the reference distribution label, screening out a local innovative resource data cluster of a second previously set dimension from the optimized local innovative resource data cluster as the optimized local innovative resource data cluster, wherein the second previously set dimension is lower than the first previously set dimension; and utilizing the optimized local innovative resource data cluster to optimize the local innovative resource data cluster of the corresponding distribution label in the reference innovative resource data.

[0039] In an independently implemented embodiment, there is no optimized constraint condition in the abnormal constraint condition in response to the reference innovation resource data, and before determining the reference distribution label in the non-optimized constraint condition, the method also includes: performing semantic mining of a previously set template on the reference innovation resource data, determining the semantic recognition condition of the previously set template in the reference innovation resource data, and using the semantic recognition condition of the previously set template as the abnormal constraint condition in the reference innovation resource data.

[0040] In an independently implemented embodiment, there is a non-optimized constraint in the abnormal constraint in response to the reference innovative resource data, and before determining the reference distribution label in the non-optimized constraint, the method further includes: transmitting the reference innovative resource data to a pre-configured second artificial intelligence thread, and determining the abnormal constraint in the reference innovative resource data based on the second artificial intelligence thread, wherein the second artificial intelligence thread is pre-configured by a plurality of innovative resource data and record data of the abnormal constraint bound one-to-one to the plurality of innovative resource data.

[0041] In a second aspect, a resource retrieval system based on big data is provided, comprising a processor and a memory communicating with each other, wherein the processor is used to retrieve a computer program from the memory and implement the above method by running the computer program.

[0042] The resource retrieval method and system based on big data provided in the embodiment of the present application obtain a configured local innovative resource data cluster based on the reference innovative resource data to be retrieved, and use the first artificial intelligence thread configured by the configured local innovative resource data cluster to perform keyword mining on the reference innovative resource data to obtain the local innovative resource data bound to the reference innovative resource data, determine the reference local innovative resource data cluster to be retrieved in the reference innovative resource data, determine the first local innovative resource data cluster bound to the reference local innovative resource data cluster in the local innovative resource data, and use the pre-configured first artificial intelligence thread to process the reference local innovative resource data cluster and the first local innovative resource data cluster to obtain the optimized local innovative resource data cluster bound to the reference local innovative resource data cluster, perform retrieval based on the local innovative resource data cluster, and obtain resource retrieval results, thereby for the first artificial intelligence thread used to optimize the reference innovative resource data, only the configured innovative resource data obtained based on the reference innovative resource data is configured, thereby improving the problem of key feature similarity between different topics of the innovative resource data, reducing interference in the reference innovative resource data and reducing the probability of abnormal reference innovative resource data label occurrence, thereby effectively improving the accuracy and completeness of the resource retrieval results. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1A flowchart of a resource retrieval method based on big data provided in an embodiment of the present application.

[0045] Figure 2 A block diagram of a resource retrieval device based on big data provided in an embodiment of the present application.

[0046] Figure 3 An architectural diagram of a big data-based resource retrieval system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0047] In order to better understand the above technical scheme, the technical scheme of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical scheme of the present application, rather than limitations on the technical scheme of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0048] See also Figure 1 , shows a resource retrieval method based on big data, which may include the technical solutions described in the following steps 11-14.

[0049] In step 11, keyword mining is performed on the reference innovation resource data to be retrieved to obtain the local innovation resource data bound to the reference innovation resource data.

[0050] In step 12, a reference local innovation resource data cluster to be retrieved in the reference innovation resource data is determined.

[0051] In step 13, in the local innovative resource data, the first local innovative resource data cluster bound to the reference local innovative resource data cluster is determined.

[0052] In step 14, the reference local innovative resource data cluster and the first local innovative resource data cluster are processed by using a pre-configured first artificial intelligence thread to obtain an optimized local innovative resource data cluster bound to the reference local innovative resource data cluster, wherein the first artificial intelligence thread is obtained by using a configured local innovative resource data cluster in advance, and the configured innovative resource data in the configured local innovative resource data cluster is obtained based on the reference innovative resource data; a search is performed based on the local innovative resource data cluster to obtain a resource search result.

[0053] In this embodiment, the reference innovation resource data may be random innovation resource data that needs to be optimized. Further, the resource search step may also include the following steps: determining keywords related to the technology in a specific industry field, specifically by using professional dictionaries, Baidu Encyclopedia and search engines to find synonyms and antonyms; using a patent search system, selecting basic keywords to search in the name field, and browsing the search results to find relevant keywords; using a patent search system, selecting basic keywords to search in the name field, and performing a technical theme analysis on the search results, and then viewing the most relevant results to find relevant keywords, etc.

[0054] In this embodiment, the reference local innovative resource data cluster is used to characterize the local innovative resource data cluster to be retrieved in the reference innovative resource data. In this embodiment, only one local innovative resource data cluster can be optimized each time, and after one local innovative resource data cluster is optimized, another local innovative resource data cluster can be optimized. That is, in this embodiment, only one reference local innovative resource data cluster to be retrieved can be determined from the reference innovative resource data each time. Of course, no less than three local innovative resource data clusters can also be optimized simultaneously, which is not limited one by one here.

[0055] In this embodiment, the reference local innovative resource data cluster and the first local innovative resource data cluster are local innovative resource data clusters with uniformly distributed labels in the reference innovative resource data and the local innovative resource data, and the dimensions of the reference local innovative resource data cluster and the first local innovative resource data cluster are consistent. For example, the dimensions of the reference local innovative resource data cluster and the first local innovative resource data cluster are both the first previously set dimensions. In this embodiment, by optimizing the reference local innovative resource data cluster in combination with the first local innovative resource data cluster, the key description content in the reference local innovative resource data cluster can be used to optimize the reference local innovative resource data cluster, thereby reducing interference.

[0056] In this embodiment, all the configured innovative resource data in the local innovative resource data cluster configured for the first artificial intelligence thread configured are obtained based on the reference innovative resource data, that is, the configuration of the first artificial intelligence thread does not depend on the innovative resource data other than the reference innovative resource data. In the prior art, when configuring an artificial intelligence thread for optimizing innovative resource data, many artificial intelligence threads configured with different innovative resource data are often used. For some designated innovative resource data to be optimized, the artificial intelligence threads configured in this way are difficult to reduce interference. In this embodiment, for the first artificial intelligence thread used to optimize the reference innovative resource data, only the configured local innovative resource data cluster is obtained based on the reference innovative resource data, and the first artificial intelligence thread is configured based on the configured local innovative resource data cluster, so that the first artificial intelligence thread is updated based on the reference innovative resource data, thereby improving the problem of similarity of key features between different themes of the innovative resource data, reducing the interference factors in the reference innovative resource data and reducing the probability of abnormal occurrence of the reference innovative resource data label, so that the accuracy and integrity of the resource retrieval results can be effectively improved.

[0057] In a possible implementation embodiment, the configured local innovative resource data cluster includes a plurality of configured innovative resource data tuples, one of the configured innovative resource data tuples includes a first resource topic and a second resource topic, wherein the first resource topic is an innovative resource data topic selected from the reference innovative resource data, the second resource topic is an innovative resource data topic selected from the local innovative resource data, and the first resource topic and the second resource topic belonging to the consistently configured innovative resource data tuple correspond to consistently selected locations; before using the pre-configured first artificial intelligence thread to process the reference local innovative resource data cluster and the first local innovative resource data cluster, the method further includes: for one of the configured innovative resource data tuples, performing abnormal behavior query processing on the first resource topic and the second resource topic to obtain a first abnormal topic bound to the first resource topic and a second abnormal topic bound to the second resource topic; transmitting the first abnormal topic and the second abnormal topic to the first artificial intelligence thread, and obtaining a first optimized topic bound to the first abnormal topic based on the first artificial intelligence thread; and configuring the first artificial intelligence thread according to the similarity between the first optimized topic and the first resource topic. Among them, the dimension of the first resource theme is lower than the dimension of the reference innovation resource data, the dimension of the second resource theme is lower than the dimension of the local innovation resource data, and the dimension of the first resource theme is consistent with the dimension of the second resource theme. The first abnormal theme is used to characterize the innovation resource data theme obtained after abnormal behavior query processing is performed on the first resource theme, and the dimension of the first abnormal theme can be consistent with the dimension of the first resource theme. The second abnormal theme is used to characterize the innovation resource data theme obtained after abnormal behavior query processing is performed on the second resource theme, and the dimension of the second abnormal theme can be consistent with the dimension of the second resource theme. The first optimized theme is used to characterize the innovation resource data theme obtained by the first artificial intelligence thread based on the optimization of the first abnormal theme and the second abnormal theme, and the dimension of the first optimized theme can be consistent with the dimension of the first abnormal theme.

[0058] In an alternative embodiment, multiple pairs of first resource themes and second resource themes can be screened out from the reference innovation resource data and the local innovation resource data, and each pair of the first resource themes and the second resource themes constitutes a configured innovation resource data tuple. Among them, the corresponding first resource themes and the second resource themes (i.e., the first resource themes and the second resource themes belonging to the consistently configured innovation resource data tuple) are local innovation resource data clusters with consistently distributed labels in the reference innovation resource data and the local innovation resource data. The dimensions of the first resource themes and the second resource themes can both be the first previously set dimensions. In an alternative embodiment, for one of the configured innovation resource data tuples, the first resource themes and the second resource themes are subjected to abnormal behavior query processing, which can refer to performing abnormal behavior query processing on pixels with consistently distributed labels in the first resource themes and the second resource themes. Among them, abnormal behavior query processing is performed on pixels with consistently distributed labels in the first resource themes and the second resource themes.

[0059] In an alternative embodiment, by implementing data expansion of the configured innovation resource data based on the reference innovation resource data, and autonomously updating the configured innovation resource data obtained by data expansion of the reference innovation resource data, the first artificial intelligence thread can be updated to the characteristics of the reference innovation resource data, thereby enabling the first artificial intelligence thread to be updated to the ability to optimize the innovation resource data for the reference innovation resource data.

[0060] In a possible real-time embodiment, the abnormal behavior query processing of the first resource topic and the second resource topic to obtain a first abnormal topic bound to the first resource topic and a second abnormal topic bound to the second resource topic includes: generating an application scenario for binding the first resource topic and the second resource topic; performing application scenario identification processing on the first resource topic using the application scenario to obtain the first abnormal topic bound to the first resource topic; performing application scenario identification processing on the second resource topic using the application scenario to obtain the second abnormal topic bound to the second resource topic.

[0061] In a possible real-time embodiment, the transmitting of the first abnormal topic and the second abnormal topic to the first artificial intelligence thread, and obtaining the first optimization topic bound to the first abnormal topic based on the first artificial intelligence thread, includes: transmitting the second abnormal topic to the first local thread of the first artificial intelligence thread, and obtaining the second optimization topic bound to the second abnormal topic based on the first local thread; transmitting the first abnormal topic and the second optimization topic to the second local thread of the first artificial intelligence thread, and obtaining the first optimization topic bound to the first abnormal topic based on the second local thread. The second optimization topic is used to characterize the innovation resource data topic obtained by the first local thread based on the optimization of the second abnormal topic, and the dimension of the second optimization topic can be consistent with the dimension of the second abnormal topic. In this example, the first artificial intelligence thread includes a first local thread and a second local thread. Of course, the two local threads can also be configured separately.

[0062] In this embodiment, the first artificial intelligence thread configured by the similarity between the first optimization topic and the first resource topic includes: the first artificial intelligence thread configured by the similarity between the first optimization topic and the first resource topic, and the similarity between the second optimization topic and the second resource topic. For example, a quantitative evaluation model can be obtained by the similarity between the first optimization topic and the first resource topic, and the similarity between the second optimization topic and the second resource topic, and the first artificial intelligence thread is configured using the quantitative evaluation model. For example, when configuring the first artificial intelligence thread, it is necessary not only to refer to the similarity between the first optimization topic output by the second local thread and the first resource topic, but also to refer to the similarity between the second optimization topic output by the first local thread and the second resource topic, so as to further improve the first artificial intelligence thread and reduce interference.

[0063] In an alternative embodiment, the first artificial intelligence thread may be configured only by the similarity between the first optimization theme and the first resource theme, without referring to the similarity between the second optimization theme and the second resource theme.

[0064] In an alternative embodiment, when configuring the first artificial intelligence thread, the reference innovation resource data and the local innovation resource data may not be screened. For example, the reference innovation resource data and the local innovation resource data may be subjected to several arbitrary abnormal behavior query processes to obtain several groups of reference innovation resource data and local innovation resource data with different abnormal constraint conditions, and the first artificial intelligence thread may be configured based on the obtained several groups of abnormal reference innovation resource data and local innovation resource data.

[0065] In an alternative embodiment, other data expansion methods for innovative resource data may be used based on the reference innovative resource data to obtain the innovative resource data configured in the configured local innovative resource data cluster.

[0066] In this embodiment, when comparing the reference local innovation resource data cluster with the optimized local innovation resource data cluster bound to it, only the local inconsistency that does not have the optimization constraint condition may be found. That is, the first artificial intelligence thread may optimize only the local inconsistency that does not have the optimization constraint condition in the reference local innovation resource data cluster. Among them, the non-optimization constraint condition is used to characterize the constraint condition that is an abnormal constraint condition in the reference innovation resource data and is not optimized, that is, the non-optimization constraint condition is used to characterize the constraint condition that is an abnormal constraint condition in the reference innovation resource data and is not optimized by the optimized local innovation resource data cluster or the optimized local innovation resource data cluster.

[0067] In a possible implementation embodiment, the method of using a pre-configured first artificial intelligence thread to process the reference local innovative resource data cluster and the first local innovative resource data cluster to obtain an optimized local innovative resource data cluster bound to the reference local innovative resource data cluster includes: transmitting the first local innovative resource data cluster to the first local thread of the pre-configured first artificial intelligence thread, and obtaining a second local innovative resource data cluster bound to the first local innovative resource data cluster based on the first local thread; and using the second local thread of the first artificial intelligence thread to process the reference local innovative resource data cluster and the second local innovative resource data cluster to obtain an optimized local innovative resource data cluster bound to the reference local innovative resource data cluster. The second local innovative resource data cluster is used to characterize the local innovative resource data cluster obtained by optimizing the first local thread to the first local innovative resource data cluster. The dimension of the second local innovative resource data cluster may be consistent with the dimension of the first local innovative resource data cluster.

[0068] In a possible real-time embodiment, the second local thread of the first artificial intelligence thread is used to process the reference local innovative resource data cluster and the second local innovative resource data cluster to obtain the optimized local innovative resource data cluster bound to the reference local innovative resource data cluster, including: performing identification processing on the reference local innovative resource data cluster to obtain the reference local innovative resource data cluster after identification processing; transmitting the reference local innovative resource data cluster and the second local innovative resource data cluster after identification processing to the second local thread of the first artificial intelligence thread, and obtaining the optimized local innovative resource data cluster bound to the reference local innovative resource data cluster based on the second local thread. , performing identification processing on the reference local innovative resource data cluster. By performing identification processing on the reference local innovative resource data cluster to obtain the reference local innovative resource data cluster after identification processing, and transmitting the reference local innovative resource data cluster and the second local innovative resource data cluster after identification processing to the second local thread, and obtaining the optimized local innovative resource data cluster bound to the reference local innovative resource data cluster based on the second local thread, interference is reduced.

[0069] In one example, the identification processing of the reference local innovative resource data cluster to obtain the reference local innovative resource data cluster after identification processing includes: determining a feature extraction set in the reference local innovative resource data cluster through the reference local innovative resource data cluster and the second local innovative resource data cluster; determining a reference feature similar to a feature of the reference local innovative resource data cluster; and extracting the reference feature into the feature extraction set to obtain the reference local innovative resource data cluster after identification processing.

[0070] In this embodiment, the feature extraction set in the reference local innovative resource data cluster is determined by using the reference local innovative resource data cluster and the second local innovative resource data cluster, including: detecting and processing the constraints to be optimized in the reference local innovative resource data cluster and the significant identification conditions in the second local innovative resource data cluster, and determining the reference optimized constraints in the reference local innovative resource data cluster, wherein the constraints to be optimized represent the constraints that are not optimized in the abnormal constraints in the reference local innovative resource data cluster; and determining the constraints other than the reference optimized constraints in the constraints to be optimized as the feature extraction set. The constraints to be optimized are sub-constraints of the reference local innovative resource data cluster, that is, the reference local innovative resource data cluster covers the constraints to be optimized, and the range of the constraints to be optimized is not greater than the range of the reference local innovative resource data cluster. Furthermore, the constraint to be optimized belongs to the non-optimized constraint in the abnormal constraint of the reference innovation resource data, that is, the constraint to be optimized is a sub-constraint of the non-optimized constraint, and the range of the constraint to be optimized is not greater than the range of the non-optimized constraint. For example, the detection and processing of the constraint to be optimized in the reference local innovation resource data cluster and the significant identification condition in the second local innovation resource data cluster. For example, by detecting and processing the constraint to be optimized in the reference local innovation resource data cluster and the significant identification condition in the second local innovation resource data cluster, the reference optimization constraint in the reference local innovation resource data cluster is determined, and the constraint other than the reference optimization constraint in the constraint to be optimized is determined as a feature extraction set, the reference optimization constraint is optimized by the second local thread, and the feature extraction set is optimized by feature extraction, so that the accuracy of the constraint optimized by the second local thread can be controlled each time.

[0071] For example, the detection and processing of the constraints to be optimized in the reference local innovation resource data cluster and the significant identification conditions in the second local innovation resource data cluster to determine the reference optimization constraints in the reference local innovation resource data cluster includes: compressing the constraints to be optimized in the reference local innovation resource data cluster to obtain the first application scenario constraints; expanding the significant identification conditions in the second local innovation resource data cluster to obtain the second application scenario constraints; and determining the fusion result of the first application scenario constraints and the second application scenario constraints as the reference optimization constraints in the reference local innovation resource data cluster.

[0072] In an alternative embodiment, the feature extraction set in the reference local innovative resource data cluster is determined by using the reference local innovative resource data cluster and the second local innovative resource data cluster, including: detecting and processing the constraints to be optimized in the reference local innovative resource data cluster to determine the reference optimized constraints in the reference local innovative resource data cluster; and determining the constraints other than the reference optimized constraints in the constraints to be optimized as the feature extraction set. For example, the constraints to be optimized in the reference local innovative resource data cluster can be compressed to obtain the reference optimized constraints in the reference local innovative resource data cluster.

[0073] In an alternative embodiment, the second local thread of the first artificial intelligence thread processes the reference local innovative resource data cluster and the second local innovative resource data cluster to obtain the optimized local innovative resource data cluster bound to the reference local innovative resource data cluster, including: transmitting the reference local innovative resource data cluster and the second local innovative resource data cluster to the second local thread, and obtaining the optimized local innovative resource data cluster bound to the reference local innovative resource data cluster based on the second local thread. For example, the reference local innovative resource data cluster may be obtained without identification processing.

[0074] In a possible implementation, the determination of the reference local innovative resource data cluster to be retrieved from the reference innovative resource data includes: in response to the presence of no optimized constraint conditions in the abnormal constraint conditions of the reference innovative resource data, determining a reference distribution label in the no optimized constraint conditions; determining the reference local innovative resource data cluster to be retrieved from the reference innovative resource data through the reference distribution label, wherein the reference local innovative resource data cluster covers the reference distribution label. Wherein, there is no optimized constraint condition in the abnormal constraint conditions of the reference innovative resource data, and the range used to characterize the constraints that are not optimized in the abnormal constraint conditions of the reference innovative resource data is not 0. In an alternative embodiment, in the case that there is no optimized constraint condition in the abnormal constraint conditions of the reference innovative resource data, a reference distribution label is determined in the no optimized constraint conditions, and the reference local innovative resource data cluster to be retrieved from the reference innovative resource data is determined based on the reference distribution label, thereby optimizing for the no optimized constraint conditions and improving the optimization effect.

[0075] In a possible real-time embodiment, the determining of the reference distribution label in the non-optimized constraint condition includes: determining a random point of significance of the non-optimized constraint condition as a reference distribution label. In this embodiment, by taking a random point of significance of the non-optimized constraint condition as a reference distribution label, and determining a reference local innovation resource data cluster to be retrieved from the reference innovation resource data based on the reference distribution label, the non-optimized constraint conditions are optimized one by one, which is conducive to further improving the optimization effect.

[0076] In this embodiment, a random point without optimization constraints may also be used as a reference distribution label.

[0077] In a possible real-time embodiment, the reference local innovation resource data cluster to be retrieved is determined from the reference innovation resource data through the reference distribution label, including: taking the reference distribution label as a benchmark, screening out the local innovation resource data cluster of the first previously set dimension from the reference innovation resource data as the reference local innovation resource data cluster to be retrieved. Screening out the reference local innovation resource data cluster through this example helps to make the range of constraints that need to be optimized and the range of constraints that do not need to be optimized in the reference local innovation resource data cluster meet the set requirements, thereby helping to improve data integrity. Of course, in other examples, the reference distribution label may not be the benchmark for the reference local innovation resource data cluster, as long as the reference local innovation resource data cluster covers the reference distribution label.

[0078] In this embodiment, after obtaining the optimized local innovative resource data cluster bound to the reference local innovative resource data cluster, the method further includes: based on the reference distribution label, screening out a local innovative resource data cluster of a second previously set dimension from the optimized local innovative resource data cluster as the optimized local innovative resource data cluster, wherein the second previously set dimension is lower than the first previously set dimension; and utilizing the optimized local innovative resource data cluster to optimize the local innovative resource data cluster of the corresponding distribution label in the reference innovative resource data.

[0079] In an alternative embodiment, in the abnormal constraint condition in response to the reference innovative resource data, there is no optimized constraint condition, and before determining the reference distribution label in the abnormal constraint condition, the method further includes: transmitting the reference innovative resource data to a pre-configured second artificial intelligence thread, and determining the abnormal constraint condition in the reference innovative resource data based on the second artificial intelligence thread, wherein the second artificial intelligence thread is configured in advance by recording data of several innovative resource data and abnormal constraint conditions bound to the several innovative resource data one by one. For example, a second artificial intelligence thread configured with several innovative resource data having different abnormal constraint conditions can be used to update the second artificial intelligence thread to the accuracy of determining the abnormal constraint condition in the innovative resource data. By transmitting the reference innovative resource data to a pre-configured second artificial intelligence thread, and determining the abnormal constraint condition in the reference innovative resource data based on the second artificial intelligence thread, the accuracy of the determined abnormal constraint condition can be improved.

[0080] The first local innovation resource data cluster can be transmitted to the first local thread, and the second local innovation resource data cluster bound to the first local innovation resource data cluster can be obtained based on the first local thread. The constraint conditions to be optimized in the reference local innovation resource data cluster can be compressed to obtain the first application scenario constraint conditions, and the significance identification conditions in the second local innovation resource data cluster can be expanded to obtain the second application scenario constraint conditions, and the fusion result of the first application scenario constraint conditions and the second application scenario constraint conditions can be determined as the reference optimization constraint conditions in the reference local innovation resource data cluster, wherein the constraint conditions to be optimized are used to characterize the constraint conditions that are not optimized in the reference local innovation resource data cluster. The constraint conditions other than the reference optimization constraint conditions in the constraint conditions to be optimized can be determined as a feature extraction set. Feature extraction is performed on the feature extraction set to obtain the reference local innovation resource data cluster after identification processing.

[0081] Based on the above, please refer to Figure 2 , a resource retrieval device 200 based on big data is provided, which is applied to a resource retrieval system based on big data, and the device comprises:

[0082] A data mining module 210 is used to perform keyword mining on the reference innovation resource data to be retrieved to obtain the local innovation resource data bound to the reference innovation resource data;

[0083] The data retrieval module 220 is used to determine a reference local innovative resource data cluster to be retrieved in the reference innovative resource data; and to determine a first local innovative resource data cluster bound to the reference local innovative resource data cluster in the local innovative resource data;

[0084] The data optimization module 230 is used to process the reference local innovative resource data cluster and the first local innovative resource data cluster by using a pre-configured first artificial intelligence thread to obtain an optimized local innovative resource data cluster bound to the reference local innovative resource data cluster, wherein the first artificial intelligence thread is obtained by using a configured local innovative resource data cluster in advance, and the configured innovative resource data in the configured local innovative resource data cluster is obtained based on the reference innovative resource data;

[0085] The result retrieval module 240 is used to perform a search based on the local innovative resource data cluster to obtain a resource search result.

[0086] Based on the above, please refer to Figure 3 , shows a resource retrieval system 300 based on big data, including a processor 310 and a memory 320 that communicate with each other, and the processor 310 is used to read and execute a computer program from the memory 320 to implement the above method.

[0087] Based on the above, a computer-readable storage medium is also provided, on which a computer program stored implements the above method when running.

[0088] In summary, based on the above scheme, a local innovative resource data cluster is configured based on the reference innovative resource data to be retrieved, and the first artificial intelligence thread configured by the configured local innovative resource data cluster is used to perform keyword mining on the reference innovative resource data to obtain the local innovative resource data bound to the reference innovative resource data, and determine the reference local innovative resource data cluster to be retrieved in the reference innovative resource data. In the local innovative resource data, the first local innovative resource data cluster bound to the reference local innovative resource data cluster is determined, and the reference local innovative resource data cluster and the first local innovative resource data cluster are processed by the pre-configured first artificial intelligence thread to obtain the optimized local innovative resource data cluster bound to the reference local innovative resource data cluster, and the search is performed based on the local innovative resource data cluster to obtain the resource search result. Therefore, for the first artificial intelligence thread used to optimize the reference innovative resource data, only the configured innovative resource data obtained based on the reference innovative resource data is used for configuration, thereby improving the problem of key feature similarity between different topics of the innovative resource data, reducing the interference in the reference innovative resource data and reducing the probability of abnormality in the reference innovative resource data label, so that the accuracy and completeness of the resource search result can be effectively improved.

[0089] It should be understood that the system and its modules shown above can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art will understand that the above methods and systems can be implemented using computer executable instructions and / or included in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of the present application can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software such as executed by various types of processors, and can also be implemented by a combination of the above hardware circuits and software (e.g., firmware).

[0090] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced may be any one or a combination of the above, or any other beneficial effects that may be obtained.

[0091] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of the present application. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements and amendments to the present application. Such modifications, improvements and amendments are suggested in the present application, so such modifications, improvements and amendments still belong to the spirit and scope of the exemplary embodiments of the present application.

[0092] At the same time, the present application uses specific words to describe the embodiments of the present application. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.

[0093] In addition, it will be appreciated by those skilled in the art that various aspects of the present application may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may all be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of the present application may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.

[0094] A computer storage medium may include a propagated data signal containing computer program code, for example, in baseband or as part of a carrier wave. The propagated signal may be in a variety of forms, including electromagnetic, optical, etc., or a suitable combination. A computer storage medium may be any computer-readable medium other than a computer-readable storage medium, which can be connected to an instruction execution system, device or apparatus to communicate, propagate or transmit the program for use. The program code on the computer storage medium may be transmitted via any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

[0095] The computer program codes required for the operation of each part of the present application can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., conventional procedural programming languages ​​such as C language, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby and Groovy, or other programming languages, etc. The program code can be run entirely on the user's computer, or run on the user's computer as an independent software package, or run partially on the user's computer and partially on a remote computer, or run entirely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer through any network form, such as a local area network (LAN) or a wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as software as a service (SaaS).

[0096] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this application, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this application. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this application. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0097] Similarly, it should be noted that in order to simplify the description of the disclosure of this application and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this application, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the object of this application are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0098] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used for the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers allow adaptive changes. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which can be changed according to the required characteristics of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining the digits. Although the numerical domains and parameters used to confirm the breadth of the range in some embodiments of the present application are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0099] Each patent, patent application, patent application disclosure, and other materials, such as articles, books, instructions, publications, documents, etc., cited in this application are hereby incorporated by reference in their entirety. Except for application history documents that are inconsistent with or conflicting with the content of this application, documents that limit the broadest scope of the claims of this application (currently or later attached to this application) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the attached materials of this application are inconsistent or conflicting with the content described in this application, the descriptions, definitions, and / or use of terms in this application shall prevail.

[0100] Finally, it should be understood that the embodiments described in this application are only used to illustrate the principles of the embodiments of the present application. Other variations may also fall within the scope of the present application. Therefore, as an example and not a limitation, the alternative configurations of the embodiments of the present application may be considered to be consistent with the teachings of the present application. Accordingly, the embodiments of the present application are not limited to the embodiments explicitly introduced and described in the present application.

[0101] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A resource retrieval method based on big data, characterized in that: Applied to a resource retrieval system, the method at least comprises: Perform keyword mining on the reference innovation resource data to be retrieved to obtain local innovation resource data bound to the reference innovation resource data; Determine a reference local innovative resource data cluster to be retrieved in the reference innovative resource data; determine a first local innovative resource data cluster bound to the reference local innovative resource data cluster in the local innovative resource data; Processing the reference local innovative resource data cluster and the first local innovative resource data cluster using a pre-configured first artificial intelligence thread to obtain an optimized local innovative resource data cluster bound to the reference local innovative resource data cluster, wherein the first artificial intelligence thread is obtained by configuring the configured local innovative resource data cluster in advance, and the configured innovative resource data in the configured local innovative resource data cluster is obtained based on the reference innovative resource data; Performing a search based on the local innovation resource data cluster to obtain a resource search result; The method of processing the reference local innovative resource data cluster and the first local innovative resource data cluster by using a pre-configured first artificial intelligence thread to obtain an optimized local innovative resource data cluster bound to the reference local innovative resource data cluster includes: The first local innovative resource data cluster is transmitted to a first local thread of a pre-configured first artificial intelligence thread, and a second local innovative resource data cluster bound to the first local innovative resource data cluster is obtained based on the first local thread; Using the second local thread of the first artificial intelligence thread to process the reference local innovative resource data cluster and the second local innovative resource data cluster to obtain an optimized local innovative resource data cluster bound to the reference local innovative resource data cluster; The method of using the second local thread of the first artificial intelligence thread to process the reference local innovative resource data cluster and the second local innovative resource data cluster to obtain an optimized local innovative resource data cluster bound to the reference local innovative resource data cluster includes: Performing identification processing on the reference local innovative resource data cluster to obtain an identified reference local innovative resource data cluster; Transmitting the reference local innovative resource data cluster and the second local innovative resource data cluster after the identification processing to the second local thread of the first artificial intelligence thread, and obtaining the optimized local innovative resource data cluster bound to the reference local innovative resource data cluster based on the second local thread; The step of performing identification processing on the reference local innovative resource data cluster to obtain the reference local innovative resource data cluster after identification processing comprises: Determining a feature extraction set in the reference local innovative resource data cluster through the reference local innovative resource data cluster and the second local innovative resource data cluster; determining reference features similar to features of the reference local innovation resource data cluster; Extracting the reference features into the feature extraction set to obtain a reference local innovation resource data cluster after identification processing; The step of determining a feature extraction set in the reference local innovative resource data cluster by using the reference local innovative resource data cluster and the second local innovative resource data cluster comprises: Detect and process the constraints to be optimized in the reference local innovative resource data cluster and the significant identification conditions in the second local innovative resource data cluster to determine the reference optimized constraints in the reference local innovative resource data cluster, wherein the constraints to be optimized represent the constraints that are not optimized among the abnormal constraints in the reference local innovative resource data cluster; determine the constraints other than the reference optimized constraints in the constraints to be optimized as a feature extraction set; Among them, the detection and processing of the constraints to be optimized in the reference local innovation resource data cluster and the significant identification conditions in the second local innovation resource data cluster to determine the reference optimization constraints in the reference local innovation resource data cluster includes: compressing the constraints to be optimized in the reference local innovation resource data cluster to obtain the first application scenario constraints; expanding the significant identification conditions in the second local innovation resource data cluster to obtain the second application scenario constraints; and determining the fusion result of the first application scenario constraints and the second application scenario constraints as the reference optimization constraints in the reference local innovation resource data cluster; Wherein, the determining of the reference local innovative resource data cluster to be retrieved in the reference innovative resource data comprises: in response to the presence of no optimized constraint conditions in the abnormal constraint conditions of the reference innovative resource data, determining a reference distribution label in the no optimized constraint conditions; determining the reference local innovative resource data cluster to be retrieved from the reference innovative resource data through the reference distribution label, wherein the reference local innovative resource data cluster covers the reference distribution label; Wherein, determining the reference distribution label in the absence of optimization constraints comprises: determining a significant random point in the absence of optimization constraints as the reference distribution label; The step of determining the reference local innovative resource data cluster to be retrieved from the reference innovative resource data by using the reference distribution tag comprises: taking the reference distribution tag as a reference, selecting the local innovative resource data cluster of the first previously set dimension from the reference innovative resource data as the reference local innovative resource data cluster to be retrieved; Wherein, after obtaining the optimized local innovative resource data cluster bound to the reference local innovative resource data cluster, the method further comprises: taking the reference distribution label as a benchmark, screening out a local innovative resource data cluster of a second previously set dimension from the optimized local innovative resource data cluster as the optimized local innovative resource data cluster, wherein the second previously set dimension is lower than the first previously set dimension; and optimizing the local innovative resource data cluster of the corresponding distribution label in the reference innovative resource data using the optimized local innovative resource data cluster; Wherein, in the abnormal constraint condition in response to the reference innovation resource data, there is no optimization constraint condition, and before determining the reference distribution label in the no optimization constraint condition, the method further comprises: performing semantic mining of a previously set template on the reference innovation resource data, determining a semantic recognition condition of the previously set template in the reference innovation resource data, and using the semantic recognition condition of the previously set template as the abnormal constraint condition in the reference innovation resource data; Wherein, there is no optimized constraint condition in the abnormal constraint condition in response to the reference innovative resource data, and before determining the reference distribution label in the non-optimized constraint condition, the method further includes: transmitting the reference innovative resource data to a pre-configured second artificial intelligence thread, and determining the abnormal constraint condition in the reference innovative resource data based on the second artificial intelligence thread, wherein the second artificial intelligence thread is pre-configured by a plurality of innovative resource data and record data of the abnormal constraint conditions bound one by one to the plurality of innovative resource data.

2. The method according to claim 1, characterized in that The configured local innovative resource data cluster includes a plurality of configured innovative resource data tuples, one of the plurality of configured innovative resource data tuples includes a first resource theme and a second resource theme, wherein the first resource theme is an innovative resource data theme screened from the reference innovative resource data, the second resource theme is an innovative resource data theme screened from the local innovative resource data, and the first resource theme and the second resource theme belonging to the consistently configured innovative resource data tuple correspond to consistently screened locations; before the reference local innovative resource data cluster and the first local innovative resource data cluster are processed by using a pre-configured first artificial intelligence thread, the method further includes: For one of the several configured innovation resource data tuples, performing abnormal behavior query processing on the first resource topic and the second resource topic to obtain a first abnormal topic bound to the first resource topic and a second abnormal topic bound to the second resource topic; Transmitting the first abnormal topic and the second abnormal topic to the first artificial intelligence thread, and obtaining a first optimized topic bound to the first abnormal topic based on the first artificial intelligence thread; The first artificial intelligence thread is configured based on the similarity between the first optimization topic and the first resource topic.

3. The method according to claim 2, characterized in that The performing abnormal behavior query processing on the first resource subject and the second resource subject to obtain a first abnormal subject bound to the first resource subject and a second abnormal subject bound to the second resource subject includes: Generate an application scenario for binding the first resource theme and the second resource theme; Performing application scenario identification processing on the first resource theme by using the application scenario to obtain a first abnormal theme bound to the first resource theme; The second resource theme is subjected to application scenario identification processing by utilizing the application scenario to obtain a second abnormal theme bound to the second resource theme.

4. The method according to claim 3, characterized in that The step of transmitting the first abnormal topic and the second abnormal topic to the first artificial intelligence thread, and obtaining a first optimization topic bound to the first abnormal topic based on the first artificial intelligence thread, includes: Transmitting the second abnormal topic to the first local thread of the first artificial intelligence thread, and obtaining a second optimized topic bound to the second abnormal topic based on the first local thread; The first abnormal topic and the second optimized topic are transmitted to the second local thread of the first artificial intelligence thread, and the first optimized topic bound to the first abnormal topic is obtained based on the second local thread.

5. The method according to claim 4, characterized in that The first artificial intelligence thread configured by the similarity between the first optimization theme and the first resource theme includes: the first artificial intelligence thread configured by the similarity between the first optimization theme and the first resource theme, and the similarity between the second optimization theme and the second resource theme.

6. A resource retrieval system based on big data, characterized in that: It comprises a processor and a memory communicating with each other, the processor is used to call a computer program from the memory, and implement the method according to any one of claims 1 to 5 by running the computer program.

Images