Similar event identification method and system based on distributed vector database

Abstract

The invention relates to the technical field of distributed vectors, and particularly discloses a similar event identification method and system based on a distributed vector database, and the method comprises the steps: obtaining the current event sequence information of a client, and enabling the client to carry out the serialization processing of a received current report event, and obtaining the current event sequence information; the current event sequence information is input into a language recognition model to obtain a similar vector recognition result, and the language recognition model can perform vector information extraction on the current event sequence information to obtain current event vector information and can substitute the current event vector information into a vector database to obtain the similar vector recognition result; and storing the similar vector identification result, and returning the similar vector identification result to the client when a query instruction of the client is received. According to the similar event recognition method based on the distributed vector database, the response speed of massive events can be increased.

Description

Technical Field

[0001] The present invention relates to the field of distributed vector technology, and in particular to a similar event recognition method based on a distributed vector database, a server, a client, and a similar event recognition system based on a distributed vector database. Background Art

[0002] With the development of information technology, a large number of business events have accumulated in the fields of urban governance, emergency dispatch, etc. How to quickly discover identical or similar events in the massive amount of events, thereby reducing the repeated processing of events and improving the response speed of events has become a more urgent need. Summary of the invention

[0003] The present invention provides a similar event recognition method based on a distributed vector database, a server, a client, and a similar event recognition system based on a distributed vector database, which solve the problem of low response speed to massive events existing in the related art.

[0004] As a first aspect of the present invention, a similar event identification method based on a distributed vector database is provided, which includes: Acquire the current event sequence information of the client, wherein the client can obtain the current event sequence information after serializing the received current reported event, wherein the current reported event at least includes current event description information and current event title information; Inputting the current event sequence information into a language recognition model to obtain a similar vector recognition result, wherein the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, and can substitute the current event vector information into a vector database to obtain a similar vector recognition result, wherein the vector database includes corpus event vector information obtained after vectorization processing based on corpus event information in a corpus; The similarity vector recognition result is stored, and when a query instruction from the client is received, the similarity vector recognition result is returned to the client.

[0005] Furthermore, the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, including: Performing text vectorization processing on the current event sequence information to obtain a vectorized representation result of the current event sequence information; Vector information is extracted from the vectorized representation result of the current event sequence information to obtain the current event vector information.

[0006] Furthermore, the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, including: Performing text vectorization processing on the current event sequence information to obtain a vectorized representation result of the current event sequence information; Vector information is extracted from the vectorized representation result of the current event sequence information to obtain the current event vector information.

[0007] Further, substituting the current event vector information into a vector database to obtain a similar vector recognition result includes: Performing vector similarity calculation on the current event vector information and the corpus event vector information in the vector database to obtain a vector similarity calculation result; The vector similarity calculation result is compared with a preset similarity threshold, and the vector similarity calculation result greater than the preset similarity threshold is used as a similar vector recognition result.

[0008] Furthermore, the similar event identification method based on the distributed vector database also includes: Build a vector database based on the corpus; Create a vector database index.

[0009] Furthermore, the creating a vector database index includes: Get the cleaned corpus in real time; Read the corpus data regularly according to the preset interval to determine whether there is any update; If there is an update, the semantic vector is extracted from the corpus according to the language recognition model; The extracted meaning vector is updated to the vector database, and the ID information and event classification information in the vector database are updated.

[0010] Furthermore, the similar event identification method based on the distributed vector database also includes: Receive query instructions from the client; Extract similar events to be queried according to the query instruction; Searching the vector database according to the similar event to be queried to determine query ID information; The similarity vector recognition result is queried according to the query ID information to return the similarity vector recognition result to the client.

[0011] As another aspect of the present invention, a similar event identification method based on a distributed vector database is provided, which includes: Receive a current reported event and corpus event information in a corpus, wherein the current reported event at least includes current event description information and current event title information, and the corpus event information at least includes corpus event description information and corpus event title information; Sequencing the current reported event and the corpus event information respectively to obtain current event sequence information and corpus event sequence information; The current event sequence information and the corpus event sequence information are sent to a server, and the server can input the current event sequence information into a language recognition model to obtain a similar vector recognition result, wherein the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, and can substitute the current event vector information into a vector database to obtain a similar vector recognition result, wherein the vector database includes corpus event vector information obtained after vectorization processing according to the corpus event information in the corpus; Receive the similarity vector recognition result returned by the server.

[0012] As another aspect of the present invention, a server is provided for implementing the above-mentioned similar event identification method based on a distributed vector database, which includes: An acquisition module, used for acquiring the current event sequence information of the client, wherein the client can obtain the current event sequence information after serializing the received current reported event, wherein the current reported event at least includes current event description information and current event title information; A recognition module, configured to input the current event sequence information into a language recognition model to obtain a similar vector recognition result, wherein the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, and can substitute the current event vector information into a vector database to obtain a similar vector recognition result, wherein the vector database includes corpus event vector information obtained after vectorization processing based on corpus event information in a corpus; The storage and return module is used to store the similarity vector recognition result and return the similarity vector recognition result to the client when receiving a query instruction from the client.

[0013] As another aspect of the present invention, a client is provided for implementing the above-mentioned similar event identification method based on a distributed vector database, which includes: A first receiving module is used to receive a current reported event and corpus event information in a corpus, wherein the current reported event at least includes current event description information and current event title information, and the corpus event information at least includes corpus event description information and corpus event title information; A serialization processing module, used to perform serialization processing on the current reported event and the corpus event information respectively, to obtain current event sequence information and corpus event sequence information; A sending module, used for sending the current event sequence information and the corpus event sequence information to a server, wherein the server can input the current event sequence information into a language recognition model to obtain a similar vector recognition result, wherein the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, and can substitute the current event vector information into a vector database to obtain a similar vector recognition result, wherein the vector database includes corpus event vector information obtained after vectorization processing based on the corpus event information in the corpus; The second receiving module is used to receive the similarity vector recognition result returned by the server.

[0014] As another aspect of the present invention, a similar event recognition system based on a distributed vector database is provided, which includes: the server mentioned above and the client mentioned above, and the server is communicatively connected with the client.

[0015] The similar event recognition method based on a distributed vector database provided by the present invention obtains the current event sequence information, vectorizes the current event sequence information based on a language recognition model, and recognizes similar vectors based on a vector database to obtain similar vector recognition results, thereby reducing the repeated processing of the current massive events and improving the response speed of events. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present invention, but do not constitute a limitation of the present invention.

[0017] Figure 1 A flow chart of an implementation of a similar event identification method based on a distributed vector database provided by the present invention.

[0018] Figure 2 A flow chart of obtaining current event vector information provided by the present invention.

[0019] Figure 3 A flowchart of creating a vector database index offline provided by the present invention.

[0020] Figure 4 The query flow chart provided by the present invention.

[0021] Figure 5 A flow chart of another implementation of the similar event identification method based on a distributed vector database provided by the present invention.

[0022] Figure 6 This is a flow chart for implementing the similar event recognition system based on the distributed vector database provided by the present invention. DETAILED DESCRIPTION

[0023] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so as to describe the embodiments of the present invention described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] In this embodiment, a similar event identification method based on a distributed vector database is provided. Figure 1 is a flow chart of a similar event identification method based on a distributed vector database according to an embodiment of the present invention. Figure 1 As shown, including: S100, obtaining current event sequence information of a client, wherein the client can obtain the current event sequence information after serializing the received current reported event, wherein the current reported event at least includes current event description information and current event title information; In an embodiment of the present invention, the similar event identification method based on a distributed vector database is applied on a server side, and the server side is connected to a client side for communication. The user uploads the current reported event to the client side, and the client side can serialize the current reported event to obtain the current event sequence information.

[0027] Specifically, the current reported event at least includes current event description information and current event title information, such as emergency plan information for typhoon weather, etc. The current event title information of the current reported event can specifically be a typhoon weather emergency plan, and the current event description information of the current reported event can specifically specify a detailed emergency plan for schools, enterprises, and government departments, and clearly specify the response measures and division of responsibilities when a typhoon comes. The client can specifically serialize the current reported event through a language recognition model. In an embodiment of the present invention, the language recognition model can specifically include an NLP Transformer model, based on which the current reported event is serialized and the current event sequence information is obtained.

[0028] After receiving the current event sequence information from the client, the server can process it based on the language recognition model to obtain similarity vector recognition results.

[0029] S200, inputting the current event sequence information into a language recognition model to obtain a similar vector recognition result, wherein the language recognition model is capable of extracting vector information from the current event sequence information to obtain current event vector information, and is capable of substituting the current event vector information into a vector database to obtain a similar vector recognition result, wherein the vector database includes corpus event vector information obtained after vectorization processing based on corpus event information in a corpus; In an embodiment of the present invention, the server inputs the current event sequence information into a language recognition model, which can extract vector information from the current event sequence information to obtain the current event vector information, and perform calculations in a vector database based on the current event vector information to obtain similar vector recognition results.

[0030] It should be noted that the language recognition model in the service level can also include an NLP Transformer model, that is, the NLP Transformer model can be divided into two parts, one on the client and the other on the server, which work together to obtain similar vector recognition results.

[0031] It should also be noted that, in the embodiment of the present invention, the vector database is specifically a database specifically used to store and query vector data, which can efficiently process similarity queries of vector data. The vector database includes corpus event vector information obtained after vectorization processing of corpus event information in the corpus, that is, the vector database includes some pre-stored corpus event vector information.

[0032] S300: storing the similarity vector recognition result, and returning the similarity vector recognition result to the client when receiving a query instruction from the client.

[0033] The similarity vector recognition result is stored, and when receiving a query instruction from the client, the server can search in the stored results and return the similarity vector recognition result to the client.

[0034] The similar event recognition method based on a distributed vector database provided in an embodiment of the present invention obtains current event sequence information, vectorizes the current event sequence information based on a language recognition model, and recognizes similar vectors based on a vector database to obtain similar vector recognition results, thereby reducing the repeated processing of current massive events and improving the response speed of events.

[0035] In an embodiment of the present invention, the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, such as Figure 2 As shown, including: S210, performing text vectorization processing on the current event sequence information to obtain a vectorized representation result of the current event sequence information; It should be understood that the text vectorization of the current event sequence information can be specifically performed by first performing word segmentation on the current event sequence information, and then performing word vectorization on each word segmentation result to obtain the vectorized representation result of the current event sequence information.

[0036] Specifically, the language recognition model in the embodiment of the present invention may be a NLP Transformer model, which may perform vectorization processing on the text of the current event sequence information to obtain a vectorized representation result of the current event sequence information.

[0037] S220: Extract vector information from the vectorized representation result of the current event sequence information to obtain current event vector information.

[0038] It should be understood that vector information extraction is performed on the above vectorized representation result, so as to remove some complicated information and extract the key vector information of the current event.

[0039] In an embodiment of the present invention, substituting the current event vector information into a vector database to obtain a similar vector recognition result includes: 1) performing vector similarity calculation on the current event vector information and the corpus event vector information in the vector database to obtain a vector similarity calculation result; 2) Compare the vector similarity calculation result with a preset similarity threshold, and use the vector similarity calculation result greater than the preset similarity threshold as the similar vector recognition result.

[0040] It should be understood that the vector similarity calculation is performed between the current event vector information and the corpus event vector information in the vector database. Specifically, the vector similarity calculation can be calculated using the cosine similarity formula. The cosine similarity formula can measure the cosine value of the angle between two vectors, thereby reflecting their degree of proximity in direction. In an embodiment of the present invention, the cosine similarity calculation formula is: similarity = dot(a, b) / (norm(a) * norm(b)), where a and b represent two vectors, dot(a, b) represents their inner product, and norm(a) and norm(b) represent their norms (lengths) respectively. The norm is usually calculated using the L2 norm, that is, norm(a) = sqrt(sum(ai^2)), where ai represents the i-th element of vector a, and sum(ai^2) represents the sum of the squares of all elements of vector a.

[0041] It should be understood that the use of the cosine similarity formula can efficiently process large-scale vector data and achieve good retrieval results in practical applications.

[0042] In an embodiment of the present invention, the similar event identification method based on a distributed vector database further includes: Build a vector database based on the corpus; Create a vector database index.

[0043] It should be understood that when the client receives the content in the corpus, it can serialize it according to the language recognition model. After receiving the serialized corpus event information, the server can vectorize it and then store it in the vector database to facilitate subsequent vector similarity calculation.

[0044] In an embodiment of the present invention, creating a vector database index includes: Get the cleaned corpus in real time; Read the corpus data regularly according to the preset interval to determine whether there is any update; If there is an update, the semantic vector is extracted from the corpus according to the language recognition model; The extracted meaning vector is updated to the vector database, and the ID information and event classification information in the vector database are updated.

[0045] Specifically, Figure 3 The figure shows the process of creating a vector database index offline.

[0046] 1) Obtain the cleaned corpus from a third-party system in real time or at a scheduled time as required.

[0047] 2) Start reading the corpus data regularly every day to determine whether it has been updated.

[0048] 3) If there is no update, the process ends.

[0049] 4) If there is an update, use the Transformer model to extract the semantic vector.

[0050] 5) Update the extracted vector to the vector database.

[0051] 6) Update the vector ID and actual event classification information to the cache and provide it for query use.

[0052] 7) After the update is completed, the process ends.

[0053] In an embodiment of the present invention, the similar event identification method based on a distributed vector database further includes: Receive query instructions from the client; Extract similar events to be queried according to the query instruction; Searching the vector database according to the similar event to be queried to determine query ID information; The similarity vector recognition result is queried according to the query ID information to return the similarity vector recognition result to the client.

[0054] It should be understood that based on the created vector database index, similar vector recognition result queries can be performed for the user's query instructions. Specifically, Figure 4 The following is a schematic diagram of the main query process.

[0055] 1) The client's http interface accesses event description information.

[0056] 2) The application calls the vector extraction module to obtain vector information.

[0057] 3) Substitute the vector information into the vector database to obtain the topN similar vectors.

[0058] 4) Based on the ID fed back by the vector database, query the event type information in the cache.

[0059] In the embodiment of the present invention, the specific corpus may use a relational database as storage, such as MySQL.

[0060] In order to maximize the use of resources, the Transformer algorithm server is usually divided into a client cluster and a server cluster. The client can be a common CPU server, and the server is generally recommended to use a server with a GPU graphics card. After the split, the same model on the server can be used for other similar scenarios. When caching similar vector recognition results, Redis can generally be used, or RDB can also be used for caching.

[0061] In summary, the similar event identification method based on a distributed vector database provided in an embodiment of the present invention obtains the current event sequence information, vectorizes the current event sequence information based on a language recognition model, and identifies similar vectors based on a vector database to obtain similar vector identification results, thereby reducing the repeated processing of the current massive events and improving the response speed of events.

[0062] As another embodiment of the present invention, a similar event recognition method based on a distributed vector database is provided, wherein: Figure 5 As shown, including: S10, receiving a current reported event and corpus event information in a corpus, wherein the current reported event at least includes current event description information and current event title information, and the corpus event information at least includes corpus event description information and corpus event title information; S20, serializing the current reported event and the corpus event information respectively to obtain current event sequence information and corpus event sequence information; S30, sending the current event sequence information and the corpus event sequence information to a server, wherein the server can input the current event sequence information into a language recognition model to obtain a similar vector recognition result, wherein the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, and can substitute the current event vector information into a vector database to obtain a similar vector recognition result, wherein the vector database includes corpus event vector information obtained after vectorization processing according to the corpus event information in the corpus; S40: Receive the similarity vector recognition result returned by the server.

[0063] In an embodiment of the present invention, the similar event identification method based on a distributed vector database is applied on a client, the client is connected to a server for communication, the client performs serialization processing on a currently reported event, obtains current event sequence information, and sends the current event sequence information to the server. After receiving the current event sequence information from the client, the server can process it based on a language recognition model to obtain a similar vector recognition result.

[0064] In an embodiment of the present invention, the server inputs the current event sequence information into a language recognition model, which can extract vector information from the current event sequence information to obtain the current event vector information, and perform calculations in a vector database based on the current event vector information to obtain similar vector recognition results.

[0065] It should be noted that the language recognition model in the service level can also include an NLP Transformer model, that is, the NLP Transformer model can be divided into two parts, one on the client and the other on the server, which work together to obtain similar vector recognition results.

[0066] The similar event recognition method based on a distributed vector database provided in an embodiment of the present invention obtains current event sequence information, vectorizes the current event sequence information based on a language recognition model, and recognizes similar vectors based on a vector database to obtain similar vector recognition results, thereby reducing the repeated processing of current massive events and improving the response speed of events.

[0067] The specific working process of the similar event identification method based on a distributed vector database applied on the client provided by the present invention can refer to the description of the similar event identification method based on a distributed vector database applied on the server in the previous text, and will not be repeated here.

[0068] As another embodiment of the present invention, a server is provided for implementing the above-mentioned similar event identification method based on a distributed vector database, which includes: An acquisition module, used for acquiring the current event sequence information of the client, wherein the client can obtain the current event sequence information after serializing the received current reported event, wherein the current reported event at least includes current event description information and current event title information; A recognition module, configured to input the current event sequence information into a language recognition model to obtain a similar vector recognition result, wherein the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, and can substitute the current event vector information into a vector database to obtain a similar vector recognition result, wherein the vector database includes corpus event vector information obtained after vectorization processing based on corpus event information in a corpus; The storage and return module is used to store the similarity vector recognition result and return the similarity vector recognition result to the client when receiving a query instruction from the client.

[0069] The server provided by the present invention obtains the current event sequence information, vectorizes the current event sequence information based on a language recognition model, and identifies similar vectors based on a vector database to obtain similar vector identification results, thereby reducing the repeated processing of the current massive events and improving the response speed of events.

[0070] The specific working principle of the server provided by the present invention can refer to the description of the similar event identification method based on the distributed vector database applied on the server in the previous text, which will not be repeated here.

[0071] As another embodiment of the present invention, a client is provided for implementing the above-mentioned similar event identification method based on a distributed vector database, which includes: A first receiving module is used to receive a current reported event and corpus event information in a corpus, wherein the current reported event at least includes current event description information and current event title information, and the corpus event information at least includes corpus event description information and corpus event title information; A serialization processing module, used to perform serialization processing on the current reported event and the corpus event information respectively, to obtain current event sequence information and corpus event sequence information; A sending module, used for sending the current event sequence information and the corpus event sequence information to a server, wherein the server can input the current event sequence information into a language recognition model to obtain a similar vector recognition result, wherein the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, and can substitute the current event vector information into a vector database to obtain a similar vector recognition result, wherein the vector database includes corpus event vector information obtained after vectorization processing based on the corpus event information in the corpus; The second receiving module is used to receive the similarity vector recognition result returned by the server.

[0072] The client provided by the present invention obtains the current event sequence information, vectorizes the current event sequence information based on a language recognition model, and identifies similar vectors based on a vector database to obtain similar vector identification results, thereby reducing the repeated processing of the current massive events and improving the response speed of events.

[0073] The specific working principle of the server provided by the present invention can refer to the description of the similar event identification method based on the distributed vector database applied on the server in the previous text, which will not be repeated here.

[0074] As another embodiment of the present invention, a similar event identification system based on a distributed vector database is provided, which includes: the server mentioned above and the client mentioned above, and the server is communicatively connected with the client.

[0075] like Figure 6 As shown, a similar event recognition system based on a distributed vector database according to an embodiment of the present invention obtains current event sequence information, vectorizes the current event sequence information based on a language recognition model, and recognizes similar vectors based on a vector database to obtain similar vector recognition results, thereby reducing repeated processing of current massive events and improving event response speed.

[0076] For the description of the similar event recognition system based on a distributed vector database according to an embodiment of the present invention, reference may be made to the description of the similar event recognition method based on a distributed vector database applied on the server side, which will not be repeated here.

[0077] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A similar event recognition method based on a distributed vector database, characterized in that: include: Acquire the current event sequence information of the client, wherein the client can obtain the current event sequence information after serializing the received current reported event, wherein the current reported event at least includes current event description information and current event title information; Inputting the current event sequence information into a language recognition model to obtain a similar vector recognition result, wherein the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, and can substitute the current event vector information into a vector database to obtain a similar vector recognition result, wherein the vector database includes corpus event vector information obtained after vectorization processing based on corpus event information in a corpus; The similarity vector recognition result is stored, and when a query instruction from the client is received, the similarity vector recognition result is returned to the client.

2. The similar event identification method based on a distributed vector database according to claim 1 is characterized in that: The language recognition model can extract vector information from the current event sequence information to obtain current event vector information, including: Performing text vectorization processing on the current event sequence information to obtain a vectorized representation result of the current event sequence information; Vector information is extracted from the vectorized representation result of the current event sequence information to obtain the current event vector information.

3. The similar event identification method based on a distributed vector database according to claim 1 is characterized in that: Substituting the current event vector information into a vector database to obtain a similar vector recognition result includes: Performing vector similarity calculation on the current event vector information and the corpus event vector information in the vector database to obtain a vector similarity calculation result; The vector similarity calculation result is compared with a preset similarity threshold, and the vector similarity calculation result greater than the preset similarity threshold is used as a similar vector recognition result.

4. The similar event identification method based on a distributed vector database according to claim 1 is characterized in that: The similar event identification method based on the distributed vector database also includes: Build a vector database based on the corpus; Create a vector database index.

5. The similar event identification method based on a distributed vector database according to claim 4 is characterized in that: The creating of the vector database index comprises: Get the cleaned corpus in real time; Read the corpus data regularly according to the preset interval to determine whether there is any update; If there is an update, the semantic vector is extracted from the corpus according to the language recognition model; The extracted meaning vector is updated to the vector database, and the ID information and event classification information in the vector database are updated.

6. The similar event identification method based on a distributed vector database according to claim 1 is characterized in that: The similar event identification method based on the distributed vector database also includes: Receive query instructions from the client; Extract similar events to be queried according to the query instruction; Searching the vector database according to the similar event to be queried to determine query ID information; The similarity vector recognition result is queried according to the query ID information to return the similarity vector recognition result to the client.

7. A similar event identification method based on a distributed vector database, characterized in that: include: Receive a current reported event and corpus event information in a corpus, wherein the current reported event at least includes current event description information and current event title information, and the corpus event information at least includes corpus event description information and corpus event title information; Sequencing the current reported event and the corpus event information respectively to obtain current event sequence information and corpus event sequence information; The current event sequence information and the corpus event sequence information are sent to a server, and the server can input the current event sequence information into a language recognition model to obtain a similar vector recognition result, wherein the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, and can substitute the current event vector information into a vector database to obtain a similar vector recognition result, wherein the vector database includes corpus event vector information obtained after vectorization processing according to the corpus event information in the corpus; Receive the similarity vector recognition result returned by the server.

8. A server, used to implement the similar event identification method based on a distributed vector database as described in any one of claims 1 to 6, characterized in that: include: An acquisition module, used for acquiring the current event sequence information of the client, wherein the client can obtain the current event sequence information after serializing the received current reported event, wherein the current reported event at least includes current event description information and current event title information; A recognition module, configured to input the current event sequence information into a language recognition model to obtain a similar vector recognition result, wherein the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, and can substitute the current event vector information into a vector database to obtain a similar vector recognition result, wherein the vector database includes corpus event vector information obtained after vectorization processing based on corpus event information in a corpus; The storage and return module is used to store the similarity vector recognition result and return the similarity vector recognition result to the client when receiving a query instruction from the client.

9. A client, used to implement the similar event identification method based on a distributed vector database as claimed in claim 7, characterized in that: include: A first receiving module is used to receive a current reported event and corpus event information in a corpus, wherein the current reported event at least includes current event description information and current event title information, and the corpus event information at least includes corpus event description information and corpus event title information; A serialization processing module, used to perform serialization processing on the current reported event and the corpus event information respectively, to obtain current event sequence information and corpus event sequence information; A sending module, used for sending the current event sequence information and the corpus event sequence information to a server, wherein the server can input the current event sequence information into a language recognition model to obtain a similar vector recognition result, wherein the language recognition model can extract vector information from the current event sequence information to obtain current event vector information, and can substitute the current event vector information into a vector database to obtain a similar vector recognition result, wherein the vector database includes corpus event vector information obtained after vectorization processing based on the corpus event information in the corpus; The second receiving module is used to receive the similarity vector recognition result returned by the server.

10. A similar event recognition system based on a distributed vector database, characterized in that: include: The server as claimed in claim 8 and the client as claimed in claim 9, wherein the server is communicatively connected with the client.

Images