CN111191041A

CN111191041A - Characteristic data acquisition method, data storage method, device, equipment and medium

Info

Publication number: CN111191041A
Application number: CN201911157167.8A
Authority: CN
Inventors: 勇萌哲; 高小宏; 滕一帆; 程晶
Original assignee: Tencent Cloud Computing Beijing Co Ltd
Current assignee: Tencent Cloud Computing Beijing Co Ltd
Priority date: 2019-11-22
Filing date: 2019-11-22
Publication date: 2020-05-22

Abstract

The embodiment of the application discloses a feature data acquisition method, a data storage method, a device, equipment and a medium, and belongs to the technical field of artificial intelligence. The method comprises the following steps: acquiring a characteristic map corresponding to a target service scene; determining service features and service labels under the target service scene according to the feature map; and acquiring feature data and label data under the target service scene according to the service features and the service labels. According to the embodiment of the application, the service characteristics and the service data in the service scene are determined by acquiring the characteristic map corresponding to the service scene, and the characteristic data and the label data in the service scene are further acquired.

Description

Characteristic data acquisition method, data storage method, device, equipment and medium

Technical Field

The embodiment of the application relates to the technical field of artificial intelligence, in particular to a characteristic data acquisition method, a data storage method, a device, equipment and a medium.

Background

Before a machine learning model is trained, feature data for training the machine learning model needs to be acquired.

In the related art, technicians need to learn business related knowledge deeply and know business features and business labels in a business scene, so as to obtain feature data and label data according to the business features and the business labels. Assuming that the service scene is suspected smuggling, a technician needs to deeply learn knowledge related to the smuggling and know service features and service labels in the suspected smuggling scene, for example, the technician needs to know which relevant information of a suspect should be obtained, so as to determine whether the suspect is suspected to be smuggling according to the relevant information.

However, the service features and service labels in the related art described above need to be deeply learned by technicians, so that the feature data is acquired inefficiently.

Disclosure of Invention

The embodiment of the application provides a feature data acquisition method, a data storage method, a device, equipment and a medium, which can be used for solving the problem that in the related technology, service features and service labels need to be deeply learned by technicians, so that the feature data acquisition efficiency is low. The technical scheme is as follows:

in one aspect, an embodiment of the present application provides a method for acquiring feature data, where the method includes:

acquiring a characteristic map corresponding to a target service scene;

determining service features and service labels under the target service scene according to the feature map;

and acquiring feature data and label data under the target service scene according to the service features and the service labels.

On the other hand, an embodiment of the present application provides a data storage method, where the method includes:

receiving a characteristic map corresponding to a target service scene sent by characteristic map generation equipment;

transmitting the feature map to a feature data generation device;

receiving a machine learning model sent by the feature data generation device, wherein the machine learning model is obtained by training feature data and label data under the target service scene, the feature data and the label data are obtained based on service features and service labels determined from the feature map, and the machine learning model is used for performing label classification on the target service scene;

storing the machine learning model.

In another aspect, an embodiment of the present application provides a feature map generation method, where the method includes:

displaying an atlas generation page provided by a service platform, wherein the atlas generation page comprises a component selection area and an atlas generation area;

receiving a dragging operation corresponding to a target component from the component selection area to the map generation area, wherein the target component is a component corresponding to a service feature or a service label in a target service scene;

generating a graph block corresponding to the target component in the graph generation area; and generating a characteristic map corresponding to the target service scene according to the graph block.

In another aspect, an embodiment of the present application provides a feature data acquiring apparatus, where the apparatus includes:

the map acquisition module is used for acquiring a characteristic map corresponding to a target service scene;

the characteristic determining module is used for determining the service characteristics and the service labels under the target service scene according to the characteristic map;

and the data acquisition module is used for acquiring the characteristic data and the label data in the target service scene according to the service characteristics and the service labels.

In another aspect, an embodiment of the present application provides a data storage device, where the data storage device includes:

the map receiving module is used for receiving the feature map corresponding to the target service scene sent by the feature map generating equipment;

the map sending module is used for sending the characteristic map to the characteristic data generation equipment;

a model receiving module, configured to receive a machine learning model sent by the feature data generation device, where the machine learning model is obtained by training feature data and tag data in the target service scene, the feature data and the tag data are obtained based on service features and service tags determined from the feature map, and the machine learning model is used to perform tag classification on the target service scene;

and the model storage module is used for storing the machine learning model.

In another aspect, an embodiment of the present application provides a feature map generation apparatus, where the apparatus includes:

the system comprises a page display module, a service platform and a service module, wherein the page display module is used for displaying an atlas generation page provided by the service platform, and the atlas generation page comprises a component selection area and an atlas generation area;

an operation receiving module, configured to receive a drag operation from the component selection area to the map generation area corresponding to a target component, where the target component is a component corresponding to a service feature or a service tag in a target service scene;

the graph block generation module is used for generating a graph block corresponding to the target component in the graph generation area;

and the map generation module is used for generating a characteristic map corresponding to the target service scene according to the map block.

In yet another aspect, an embodiment of the present application provides a computer device, which includes a processor and a memory, where the memory stores at least one instruction, at least one program, a code set, or a set of instructions, and the at least one instruction, the at least one program, the code set, or the set of instructions is loaded and executed by the processor to implement the above-mentioned feature data acquiring method, or implement the above-mentioned data storing method, or implement the above-mentioned feature map generating method.

In yet another aspect, an embodiment of the present application provides a computer-readable storage medium, where at least one instruction, at least one program, a code set, or a set of instructions is stored in the storage medium, and the at least one instruction, the at least one program, the code set, or the set of instructions is loaded and executed by a processor to implement the above-mentioned feature data acquisition method, or implement the above-mentioned data storage method, or implement the above-mentioned feature map generation method.

The technical scheme provided by the embodiment of the application can bring the following beneficial effects:

by acquiring the feature map corresponding to the service scene, the service features and the service data in the service scene are determined, and further the feature data and the label data in the service scene are acquired.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a schematic illustration of an implementation environment provided by one embodiment of the present application;

FIG. 2 is a flow chart of a method for feature data acquisition provided by an embodiment of the present application;

FIG. 3 is a schematic representation of a feature map provided by one embodiment of the present application;

FIG. 4 is a schematic diagram of a property configuration page provided by one embodiment of the present application;

FIG. 5 is a schematic diagram of a property configuration page provided by another embodiment of the present application;

FIG. 6 is a flow chart of a method for feature data acquisition according to another embodiment of the present application;

FIG. 7 is a flow chart of a data storage method provided by an embodiment of the present application;

FIG. 8 is an architecture diagram of a service platform provided by an embodiment of the present application;

FIG. 9 is a flow chart of a feature map generation method provided by one embodiment of the present application;

FIG. 10 is a block diagram of a feature data acquisition device provided in one embodiment of the present application;

fig. 11 is a block diagram of a feature data acquisition device according to another embodiment of the present application;

FIG. 12 is a block diagram of a data storage device provided by one embodiment of the present application;

FIG. 13 is a block diagram of a data storage device provided in another embodiment of the present application;

FIG. 14 is a block diagram of a feature map generation apparatus provided in one embodiment of the present application;

fig. 15 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.

Artificial Intelligence (AI) is a theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human Intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the realization method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.

The artificial intelligence technology is a comprehensive subject and relates to the field of extensive technology, namely the technology of a hardware level and the technology of a software level. The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.

Computer Vision technology (CV) Computer Vision is a science for researching how to make a machine "see", and further refers to that a camera and a Computer are used to replace human eyes to perform machine Vision such as identification, tracking and measurement on a target, and further image processing is performed, so that the Computer processing becomes an image more suitable for human eyes to observe or transmitted to an instrument to detect. As a scientific discipline, computer vision research-related theories and techniques attempt to build artificial intelligence systems that can capture information from images or multidimensional data. Computer vision technologies generally include image processing, image recognition, image semantic understanding, image retrieval, OCR, video processing, video semantic understanding, video content/behavior recognition, three-dimensional object reconstruction, 3D technologies, virtual reality, augmented reality, synchronous positioning, map construction, and other technologies, and also include common biometric technologies such as face recognition and fingerprint recognition.

Natural Language Processing (NLP) is an important direction in the fields of computer science and artificial intelligence. It studies various theories and methods that enable efficient communication between humans and computers using natural language. Natural language processing is a science integrating linguistics, computer science and mathematics. Therefore, the research in this field will involve natural language, i.e. the language that people use everyday, so it is closely related to the research of linguistics. Natural language processing techniques typically include text processing, semantic understanding, machine translation, robotic question and answer, knowledge mapping, and the like.

Machine Learning (ML) is a multi-domain cross discipline, and relates to a plurality of disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and the like. The special research on how a computer simulates or realizes the learning behavior of human beings so as to acquire new knowledge or skills and reorganize the existing knowledge structure to continuously improve the performance of the computer. Machine learning is the core of artificial intelligence, is the fundamental approach for computers to have intelligence, and is applied to all fields of artificial intelligence. Machine learning and deep learning generally include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and formal education learning.

Machine learning is a necessary product of the development of artificial intelligence research to a certain stage, and aims to improve the performance of the system by means of calculation and by using experience. In a computer system, "experience" is usually in the form of "data" from which a "model" can be generated by a machine learning algorithm, i.e. by providing empirical data to a machine learning algorithm, a model can be generated based on these empirical data, which provides a corresponding judgment, i.e. a prediction, in the face of a new situation.

In the related art, the feature data is acquired by: importing a data set (e.g., a data table) containing historical data records or predictive data records; by performing various processes on the attribute information of the data records in the data set to obtain various features, a feature vector formed by the features can be used as a machine learning training sample or a machine learning prediction sample. Each data record in the data table may include a plurality of attribute information, and the characteristics may indicate the result of various field processing (or operations) such as the field itself, or a part of the field, or a combination of the fields, so as to better reflect the data distribution and the inherent association and potential meaning between the fields.

The feature map is a map showing a service scene in a graphical form, is a more formal and simplified expression form, and can include a plurality of graphic blocks. The characteristic map can be called as a knowledge map, so that more accurate information can be found for a user, more comprehensive summary can be made, and more depth-related information can be provided. The characteristic map can be artificially constructed and defined to describe the relationship among various concepts.

The scheme provided by the embodiment of the application relates to the technologies of artificial intelligence, such as computer vision technology, natural language processing, machine learning and the like, and is specifically explained by the following embodiment.

Referring to fig. 1, a schematic diagram of an implementation environment provided by an embodiment of the present application is shown. The implementation environment includes: a feature map generation apparatus 10, a feature data generation apparatus 20, and a server 30.

In the embodiment of the present application, the feature map generating device 10 refers to a device for generating a feature map, the feature map generating device 10 may be a terminal, for example, an electronic device such as a Personal Computer (PC), a tablet Computer, a mobile phone, and a smart wearable device, and the feature map generating device 10 uploads the generated feature map to the server 30.

The feature data generation device 20 refers to a device for acquiring a feature map to generate feature data, and the feature data generation device 20 may be a terminal, for example, an electronic device such as a PC, a tablet computer, a mobile phone, a smart wearable device, and the like. The feature data generation apparatus 20 may acquire the feature map generated by the feature map generation apparatus 10 from the server 30.

The server 30 may be one server, a server cluster composed of a plurality of servers, or a cloud computing center. The server 30 may communicate with the feature map generation apparatus 10 and the feature data generation apparatus 20 through a wired or wireless network. Illustratively, a service platform may be operated in the server 30, the feature map generating device 10 uploads the generated feature map to a database of the service platform, and the feature data generating device 20 obtains the feature map from the database of the service platform.

Referring to fig. 2, a flowchart of a feature data obtaining method provided in an embodiment of the present application is shown, where an execution subject of the method may be the feature data generating device shown in fig. 1, and the method may include the following steps:

step 201, obtaining a feature map corresponding to a target service scene.

The target service scenario may be any one of service scenarios, such as fraud, theft, robbery, etc. The feature maps corresponding to different service scenes are different, the feature maps refer to maps which show the service scenes in a graphical mode, and a plurality of feature maps can correspond to the same service scene.

Alternatively, the feature data generation apparatus may acquire the feature map from a database of the service platform. The feature map may be generated by a service person through a feature map generating device.

Step 202, determining service features and service labels in the target service scene according to the feature map.

In an exemplary embodiment, as shown in fig. 3, the feature map 300 includes: service profile 31, service label 32, attribute information of service profile 31, and attribute information of service label 32. The service feature 31 is used to characterize the features required in the service scenario, the service tag 32 is used to characterize the tag in the service scenario, the attribute information of the service feature 31 is used to indicate the attribute of the service feature 31, and the attribute information of the service tag 32 is used to indicate the attribute of the service tag 32. The attribute information of the service profile 31 and the attribute information of the service label 32 may include a plurality of pieces of information, each of which may correspond to one data filtering rule. For example, the service scene is an analysis of suspected smuggling, and the service characteristics 31 may include a vehicle, a vehicle track, identity information, and entry and exit information. Taking the service feature 31 as an example of the belonging vehicle, the attribute information of the belonging vehicle may include information of a belonging city, a license plate number, a brand, and the like of the belonging vehicle, the data filtering rule corresponding to the belonging city may be xx city, and the brand may be xx blue, and optionally, the attribute information of the service feature 31 further includes a creator and creation time. The attribute information of the service feature 31 may be set in the attribute configuration page as shown in fig. 4, and of course, the user may click the "modify attribute" button to modify the attribute information of a certain service feature. The service tag 32 may be suspected smuggling, and the attribute information of the service tag 32 may include tag description, for example, to determine whether a person is suspected smuggling; the attribute information of the service tag 32 may further include an influence weight and a data filtering rule corresponding to each service feature, for example, the influence weight corresponding to the belonging vehicle is 20%, the influence weight corresponding to the vehicle track is 20%, the influence weight corresponding to the identity information is 20%, the influence weight corresponding to the entry and exit information is 40%, the data filtering rule corresponding to the belonging vehicle is xx city, the data filtering rule corresponding to the vehicle track is xx map data, and the data filtering rule corresponding to the identity information is xx citizen. Optionally, the attribute information of the service tag 32 further includes a creator and a creation time. The attribute information of the service tag 32 may be set in an attribute configuration page as described in fig. 5.

For example, the service scene is used for identification of terrorist people or black people, and the service features 31 may include identity information, entry and exit information, mobile phone information, and social application information. The service label 32 may be whether it is a terrorist person or a blackish person.

Illustratively, the service features 31 include common service features and custom service features. The public service features refer to the service features which can not be modified by self-definition, and the user-defined service features refer to the service features which can be modified by self-definition.

Illustratively, the business tags 32 include public business tags and custom business tags. The public service tag is a service tag which can not be modified by self-definition, and the user-defined service tag is a service tag which can be modified by self-definition.

The public service features and the public service labels may be set by an administrator of the service platform and are authenticated authoritative service accumulations. Common users can not modify the public service features and the public service labels and can set custom service features and custom service labels.

Step 203, acquiring feature data and label data in the target service scene according to the service features and the service labels.

The feature data is data corresponding to a service feature, and the tag data is data corresponding to a service tag. The feature data generation device may obtain the feature data and the tag data from the database according to the service feature and the service tag.

The business characteristics and the business labels are used as important contents for data mining analysis, and the richness and the accuracy of the business characteristics and the business labels are directly related to the effect of data analysis application such as data portrayal and thematic analysis. The characteristic map realizes the separation of data analysis service and technology, and service personnel can output own service experience and also participate in data analysis projects to finally form precious service sediment.

In summary, in the technical scheme provided in the embodiment of the present application, by obtaining the feature map corresponding to the service scene, the service features and the service data in the service scene are determined, and then the feature data and the tag data in the service scene are obtained.

In an exemplary embodiment, as shown in fig. 6, a flowchart of a feature data obtaining method provided in another embodiment of the present application is shown, and the method may include the following steps:

step 601, obtaining a feature map corresponding to a target service scene.

Step 602, determining service features and service labels in the target service scene according to the feature map.

Step 603, according to the service characteristics and the service label, acquiring characteristic data and label data in the target service scene.

Illustratively, the feature data generating device may acquire the feature data and the tag data by:

firstly, determining a filtering rule of feature data according to the service features and the attribute information of the service features;

still taking the above example as an example for explanation, assuming that the business feature includes the belonging vehicle, and the attribute information of the belonging vehicle includes the city, the license plate number, and the brand of the belonging vehicle, it may be determined that the filtering rule of the feature data is a certain city and the brand is a vehicle of xx blue.

Secondly, acquiring feature data which accords with the filtering rule of the feature data;

the characteristic data generating device can screen out a vehicle with a city and a brand xx blue from a database as the characteristic data of the vehicle.

Thirdly, determining a filtering rule of the label data according to the service label and the attribute information of the service label;

still taking the above example as an example for explanation, the service tag includes suspected smuggling, the attribute information of suspected smuggling includes that the influence weight corresponding to the affiliated vehicle is 20%, the influence weight corresponding to the vehicle track is 20%, the influence weight corresponding to the identity information is 20%, the influence weight corresponding to the entry and exit information is 40%, the data filtering rule corresponding to the affiliated vehicle is xx city, the data filtering rule corresponding to the vehicle track is xx map data, and the data filtering rule corresponding to the identity information is xx citizens.

Fourthly, the label data which accords with the filtering rule of the label data is obtained.

The characteristic data generation equipment can screen out the data suspected of smuggling from the database as the label data.

And step 604, generating a training sample of the machine learning model according to the feature data and the label data.

Machine learning is a method that can give machine learning ability so as to make it complete functions that cannot be completed by direct programming, and is a method that trains a model by using data and then uses model prediction. Training samples refer to data that train a machine learning model. The feature data generation device may compose the feature data and the label data into a training sample.

Step 605, training the machine learning model by using the training samples.

In the embodiment of the application, the machine learning model is used for performing label classification on the target business scene. For example, the feature data generation device may train the machine learning model a preset number of times using the training samples. The feature data generation device may train the machine learning model according to the label classification and the service label predicted by the machine learning model.

In summary, in the technical solution provided in the embodiment of the present application, the training sample of the machine learning model is generated according to the feature data and the label data, and the machine learning model is trained by using the training sample, because the training sample of the machine learning model is determined according to the service features and the service labels included in the feature map generated by the service personnel, and the service experience of the service personnel is fused, the generated training sample is more accurate, and thus the training of the machine learning model is more accurate. The feature data determines the upper limit of machine learning, and the machine learning model and algorithm only approximate the upper limit, so accurate training samples can make machine learning more optimal.

In an exemplary embodiment, after the machine learning model is trained, the feature data generation apparatus may adjust the feature map according to the training result of the machine learning model by:

receiving a graph block deleting instruction corresponding to the feature map; and deleting the first graphic block in the feature map according to the graphic block deleting instruction.

Receiving a graphic block adding instruction corresponding to the characteristic map; and adding a second graphic block in the feature map according to the graphic block adding instruction.

Receiving a graphic block modification instruction corresponding to the characteristic map; and modifying the third graphic block in the feature map according to the graphic block modification instruction.

The graph block deletion instruction is used for indicating to delete the first graph block in the feature map. The graphic block adding instruction is used for indicating that a second graphic block is added in the feature map. The graphics block modification instruction is to instruct modification of a third graphics block in the feature map.

The feature data generation equipment can delete the graphic blocks in the feature map and can add the graphic blocks in the feature map; or, the feature data generation device may delete the graphic blocks in the feature map, or modify the graphic blocks in the feature map; or, the feature data generation device may add a graphic block in the feature map, or modify a graphic block in the feature map; alternatively, the skilled person may perform deletion, addition and modification operations on the graph blocks in the feature map.

The graphic block may be a graphic block corresponding to the service feature, or a graphic block corresponding to the service tag. If the label classification indicated by the training result of the machine learning model is inconsistent with the service label, the feature data generation device may delete the service feature inconsistent with the service label and add the graphic block corresponding to the service feature corresponding to the service label. The technical personnel can input the newly found service characteristics and service labels into the characteristic map through technical means such as machine learning and the like, and can also verify whether the characteristics and labels deposited by the service personnel according to service experience are accurate through data and algorithms so as to reflow and optimize the characteristic map.

In an exemplary embodiment, after the feature data generation device trains the machine learning model with the training samples, the following steps may be further performed:

1. generating an AI (Artificial Intelligence) template according to the machine learning model and the characteristic map;

2. and storing the AI template into a database of the service platform.

The feature map and the correspondingly trained machine learning model form an AI template, a high-depth artificial intelligence technology is not required to be mastered, business personnel with a certain artificial intelligence technology basis can quickly carry out AI related work according to the AI template, project implementation cost is saved, and project implementation quality is guaranteed.

In summary, in the technical solution provided in the embodiment of the present application, the feature map may be optimized by performing deletion, addition, or modification operations on the feature map, so that the finally generated feature map is more optimal.

In addition, an AI template is generated according to the machine learning model and the characteristic map, and the AI template is stored in a database of the service platform, so that the project development is simpler and more efficient, and the project implementation cost is saved.

Referring to fig. 7, a flowchart of a data storage method according to an embodiment of the present application is shown. The execution subject of the method may be the server in fig. 1, and the method may include the following steps:

step 701, receiving a feature map corresponding to a target service scene sent by a feature map generating device.

And the service personnel generates a characteristic map corresponding to the target service scene in the service platform through the characteristic map generating equipment and stores the characteristic map into a database of the service platform. The technical architecture of the service platform provided in the embodiment of the present application may be as shown in fig. 8, where the service platform may be constructed by using a big data correlation technique, and a platform bottom layer architecture (database) adopts a hadoop (distributed) system infrastructure, including: hbase (distributed database), hive (data warehouse tool), hdfs (Hadoop partitioned File System). The program structure adopts MVC (Model View controller) three-layer structure system, the service layer includes: data source management, file data management, data flow management, feature map management, and the like. The single sign-on of the user is realized by using a cas (central authentication service) technology. The data layer includes service data BEAN and service data DAO. The service data BEAN is a reusable component written in java language. The service Data DAO (Data Access Object) is used to isolate different databases. cas-client (customer in central authentication service) supports a very large number of clients, and cas-server (server in central authentication service) is responsible for authentication work of users. The feature map generation equipment can log in a service platform to realize uploading of the feature map; the characteristic data generating equipment can log in the service platform to realize the acquisition of the characteristic map. The operation environment of the service platform can comprise a cloud host, a local server and a third-party virtual host.

Step 702, sending the feature map to the feature data generation device.

The service platform may send the feature map to the feature data generating device, for example, the feature data generating device may send a map obtaining request to the server, where the map obtaining request is used to request to obtain the feature map, and the server sends the feature map from the database of the service platform to the feature data generating device when receiving the map obtaining request.

And step 703, receiving the machine learning model sent by the feature data generation device.

In the embodiment of the application, the machine learning model is obtained by training feature data and label data in a target service scene, the feature data and the label data are obtained based on service features and service labels determined from a feature map, and the machine learning model is used for performing label classification on the target service scene.

For the description of how the feature data generating device trains the machine learning model, reference may be made to the above embodiments, which are not described herein again. After the feature data generation device has trained the machine learning model, the machine learning model may be sent to a database of the service platform.

Step 704, store the machine learning model.

The server, after receiving the machine learning model, stores the machine learning model.

In an exemplary embodiment, the feature map includes: the service characteristics, the service labels, the attribute information of the service characteristics and the attribute information of the service labels; in the embodiment of the present application, the attribute information of the service feature is used to indicate an attribute of the service feature, and the attribute information of the service tag is used to indicate an attribute of the service tag.

For the description of the characteristic map, reference may be made to the above examples, which are not repeated herein.

Illustratively, the server may further receive the modified feature map transmitted by the feature data generation device; storing the modified feature map.

In the embodiment of the application, the modified feature map is obtained by modifying the feature map according to the training result of the machine learning model, and the feature data generating device can modify the feature map and send the modified feature map to the server, so that the server stores the modified feature map.

Illustratively, the server may also receive an AI template sent by the feature data generating device; the AI template is stored.

In the embodiment of the application, the AI template is generated according to a machine learning model and a feature map. The feature data generation equipment generates the AI template according to the machine learning model and the feature map, and sends the AI template to the server so as to realize the sharing of the AI template, thereby being beneficial to business personnel to quickly develop AI related development work according to the AI template, saving project implementation cost and ensuring project implementation quality.

After the machine learning model is stored in the service platform, service personnel can acquire the machine learning model from the service platform through own equipment. Assuming that the service personnel are public security personnel, the public security personnel are facing the criminal suspects of smuggling together, the criminal suspects comprise a suspect A, a suspect B, a suspect C and a suspect D, and the public security personnel need to determine which are the suspects of smuggling and which are innocent personnel. At this time, the policeman can obtain a machine learning model related to the suspected smuggle from the service platform, and inputs the vehicles, vehicle tracks, identity information and entry and exit information of the suspect a, the suspect B, the suspect C and the suspect D into the machine learning model, and the machine learning model outputs that the suspect a is the suspected smuggle, the suspect B is an innocent person, the suspect C is an innocent person, and the suspect D is the suspected smuggle, and the machine learning model identifies and analyzes the suspected smuggle through the machine learning model, so that more reference information can be provided for the public security officer to solve the case, and the difficulty of breaking the case is reduced.

In summary, in the technical solution provided in the embodiment of the present application, by receiving the feature map sent by the feature map generation device, sending the feature map to the feature data generation device, and then receiving the machine learning model sent by the feature data generation device, sharing of the feature map and the machine learning model can be achieved, and the efficiency of obtaining feature data is improved, thereby improving the efficiency of data analysis.

In addition, the server can also receive the modified feature map and the AI template sent by the feature data generation equipment, so that information sharing can be realized, business personnel can rapidly carry out AI related work according to the information, project implementation cost is saved, and project implementation quality is ensured.

Referring to fig. 9, a flowchart of a feature map generation method provided in an embodiment of the present application is shown, where an execution subject of the method may be the feature map generation apparatus shown in fig. 1, and the method may include the following steps:

and step 901, displaying an atlas generation page provided by the service platform.

In the present embodiment, as shown in fig. 3, the atlas generation page 400 includes a component selection area 401 and an atlas generation area 402. The component selection area 401 refers to an area for a user to select a component, and the atlas generation area 402 refers to an area for generating the feature atlas 300.

Step 902, receiving a drag operation corresponding to a target component from a component selection area to a map generation area.

In the embodiment of the present application, a target component refers to a component corresponding to a service feature or a service tag in a target service scenario.

And step 903, generating a graph block corresponding to the target component in the graph generation area.

Optionally, a graph block corresponding to the target component is generated in the graph generation area through the service platform.

Still taking the service scene as an example of suspected smuggling analysis for description, the feature map generation device receives a drag operation from the component selection area 401 to the map generation area 402 corresponding to the component corresponding to the belonging vehicle, and generates a map block corresponding to the belonging vehicle in the map generation area 402. Accordingly, the feature map generation device receives a drag operation from the component selection area 401 to the map generation area 402 corresponding to the component corresponding to the suspected smuggling, and generates a map block corresponding to the suspected smuggling in the map generation area 402. The feature map generation device generates a map block corresponding to the vehicle trajectory, the identity information, and the entry and exit information in the map generation area 402 according to the above-described procedure.

And 904, generating a characteristic map corresponding to the target service scene according to the map block.

The feature map generation device generates a feature map 300 corresponding to suspected smuggling analysis according to the corresponding map blocks of the vehicle, the vehicle track, the identity information, the entry and exit information and the suspected smuggling.

To sum up, in the technical solution provided in the embodiment of the present application, a graph block corresponding to a component is generated in a graph generation region by receiving a drag operation corresponding to the component from a component selection region to a graph generation region, so that a feature graph corresponding to a service scene is generated according to the graph block, and the generation process of the feature graph is simple and efficient in operation.

The following are embodiments of the apparatus of the present application that may be used to perform embodiments of the method of the present application. For details which are not disclosed in the embodiments of the apparatus of the present application, reference is made to the embodiments of the method of the present application.

Referring to fig. 10, a block diagram of a feature data acquiring apparatus according to an embodiment of the present application is shown. The device has the function of realizing the characteristic data acquisition method example, and the function can be realized by hardware or by hardware executing corresponding software. The apparatus 1000 may include: an atlas acquisition module 1010, a feature determination module 1020, and a data acquisition module 1030.

The map obtaining module 1010 is configured to obtain a feature map corresponding to the target service scene.

A feature determining module 1020, configured to determine, according to the feature map, a service feature and a service tag in the target service scenario.

A data obtaining module 1030, configured to obtain feature data and tag data in the target service scenario according to the service feature and the service tag.

In an exemplary embodiment, the feature map includes: the service feature, the service label, the attribute information of the service feature and the attribute information of the service label;

the attribute information of the service feature is used for indicating the attribute of the service feature, and the attribute information of the service tag is used for indicating the attribute of the service tag.

In an exemplary embodiment, the data obtaining module 1030 is configured to:

determining a filtering rule of the feature data according to the service features and the attribute information of the service features;

acquiring the feature data which accords with the filtering rule of the feature data;

determining a filtering rule of the label data according to the service label and the attribute information of the service label;

and acquiring the label which accords with the filtering rule of the label data.

In an exemplary embodiment, the service features include common service features and custom service features;

the public service features refer to service features which cannot be modified by self-definition, and the user-defined service features refer to service features which can be modified by self-definition.

In an exemplary embodiment, as illustrated in fig. 11, the apparatus 1000 further includes: a sample generation module 1040 and a model training module 1050.

And the sample generating module 1040 is configured to generate a training sample of the machine learning model according to the feature data and the label data.

And a model training module 1050 configured to train the machine learning model with the training samples, where the machine learning model is configured to perform label classification on the target service scene.

In an exemplary embodiment, the apparatus 1000 further comprises: a graphics modification module 1060.

The graphics modification module 1060 is configured to:

receiving a graph block deleting instruction corresponding to the feature map, wherein the graph block deleting instruction is used for indicating to delete a first graph block in the feature map; deleting the first graphic block in the feature map according to the graphic block deleting instruction;

and/or the presence of a gas in the gas,

receiving a graphic block adding instruction corresponding to the feature map, wherein the graphic block adding instruction is used for indicating that a second graphic block is added in the feature map; adding the second graphic block in the feature map according to the graphic block adding instruction;

and/or the presence of a gas in the gas,

receiving a graphic block modification instruction corresponding to the feature map, the graphic block modification instruction being for instructing modification of a third graphic block in the feature map; and modifying the third graphic block in the feature map according to the graphic block modification instruction.

In an exemplary embodiment, the apparatus 1000 further comprises: a template generation module 1070 and a template storage module 1080.

The template generation module 1070 is configured to generate an artificial intelligence AI template according to the machine learning model and the feature map;

and a template storage module 1080, configured to store the AI template in a database of the service platform.

Referring to fig. 12, a block diagram of a data storage device according to an embodiment of the present application is shown. The device has the functions of realizing the data storage method examples, and the functions can be realized by hardware or by hardware executing corresponding software. The apparatus 1200 may include: an atlas receiving module 1210, an atlas sending module 1220, a model receiving module 1230, and a model storage module 1240.

The map receiving module 1210 is configured to receive a feature map corresponding to a target service scene sent by a feature map generating device.

A map sending module 1220, configured to send the feature map to a feature data generation device.

A model receiving module 1230, configured to receive a machine learning model sent by the feature data generation device, where the machine learning model is obtained by training feature data and tag data in the target service scene, the feature data and the tag data are obtained based on service features and service tags determined from the feature map, and the machine learning model is used to perform tag classification on the target service scene.

A model storage module 1240 for storing the machine learning model.

In an exemplary embodiment, as shown in fig. 13, the apparatus 1200 further includes: the map storage module 1250.

The map receiving module 1210 is further configured to receive a modified feature map sent by the feature data generating device, where the modified feature map is obtained by modifying the feature map according to a training result of the machine learning model.

A map storage module 1250 configured to store the modified feature map.

In an exemplary embodiment, the apparatus further comprises: template receiving module 1260 and template storage module 1270.

A template receiving module 1260, configured to receive the artificial intelligence AI template sent by the feature data generating device, where the AI template is generated according to the machine learning model and the feature map.

A template storage module 1270, configured to store the AI template.

Referring to fig. 14, a block diagram of a feature map generation apparatus according to an embodiment of the present application is shown. The device has the function of realizing the above characteristic map generation method, and the function can be realized by hardware or by hardware executing corresponding software. The apparatus 1400 may include: a page display module 1410, an operation receiving module 1420, a graphic block generation module 1430, and a map generation module 1440.

And the page display module 1410 is configured to display an atlas generation page provided by the service platform, where the atlas generation page includes a component selection area and an atlas generation area.

An operation receiving module 1420, configured to receive a drag operation from the component selection area to the map generation area corresponding to a target component, where the target component is a component corresponding to a service feature or a service tag in a target service scene.

And the graph block generation module 1430 is configured to generate a graph block corresponding to the target component in the graph generation area.

And the map generating module 1440 is configured to generate a feature map corresponding to the target service scenario according to the map block.

It should be noted that, when the apparatus provided in the foregoing embodiment implements the functions thereof, only the division of the functional modules is illustrated, and in practical applications, the functions may be distributed by different functional modules according to needs, that is, the content structure of the device may be divided into different functional modules to implement all or part of the functions described above. In addition, the apparatus and method embodiments provided by the above embodiments belong to the same concept, and specific implementation processes thereof are described in the method embodiments for details, which are not described herein again.

Referring to fig. 15, a schematic structural diagram of a computer device 1500 according to an embodiment of the present application is shown. The computer device refers to an electronic device with computing and processing capabilities, and the computer device may be a terminal or a server. The computer device 1500 may be used to implement the methods provided in the embodiments described above. Specifically, the method comprises the following steps:

the computer device 1500 includes a Central Processing Unit (CPU) 1501, a system Memory 1504 including a RAM (Random Access Memory) 1502 and a ROM (Read-Only Memory) 1503, and a system bus 1505 connecting the system Memory 1504 and the Central Processing Unit 1501. The computer device 1500 also includes a basic Input/Output system (I/O system) 1506 for facilitating information transfer between devices within the computer, and a mass storage device 1507 for storing an operating system 1513, application programs 1514 and other program modules 1515.

The basic input/output system 1506 includes a display 1508 for displaying information and an input device 1509 such as a mouse, keyboard, etc. for a user to input information. Wherein the display 1508 and the input device 1509 are connected to the central processing unit 1501 via an input output controller 1510 connected to the system bus 1505. The basic input/output system 1506 may also include an input/output controller 1510 for receiving and processing input from a number of other devices, such as a keyboard, mouse, or electronic stylus. Similarly, the input-output controller 1510 also provides output to a display screen, a printer, or other type of output device.

The mass storage device 1507 is connected to the central processing unit 1501 through a mass storage controller (not shown) connected to the system bus 1505. The mass storage device 1507 and its associated computer-readable media provide non-volatile storage for the computer device 1500. That is, the mass storage device 1507 may include a computer-readable medium (not shown) such as a hard disk or a CD-ROM (Compact disk Read-Only Memory) drive.

Without loss of generality, the computer-readable media may comprise computer storage media and communication media. Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes RAM, ROM, EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), Flash Memory (Flash Memory) or other solid state Memory technology, CD-ROM, DVD (Digital versatile disk) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices. Of course, those skilled in the art will appreciate that the computer storage media is not limited to the foregoing. The system memory 1504 and mass storage device 1507 described above may be collectively referred to as memory.

According to various embodiments of the present application, the computer device 1500 may also operate as a remote computer connected to a network via a network, such as the Internet. That is, the computer device 1500 may be connected to the network 1512 through the network interface unit 1511 connected to the system bus 1505 or may be connected to other types of networks or remote computer systems (not shown) using the network interface unit 1511.

The memory also includes one or more programs stored in the memory and configured to be executed by one or more processors. The one or more programs contain instructions for implementing the methods described above.

In an exemplary embodiment, a computer device is also provided that includes a processor and a memory having at least one instruction, at least one program, set of codes, or set of instructions stored therein. The at least one instruction, at least one program, set of codes, or set of instructions is configured to be executed by one or more processors to implement the above-described method of feature data acquisition.

In an exemplary embodiment, a computer device is also provided that includes a processor and a memory having at least one instruction, at least one program, set of codes, or set of instructions stored therein. The at least one instruction, at least one program, set of code or set of instructions is configured to be executed by one or more processors to implement the above-described data storage methods.

In an exemplary embodiment, a computer device is also provided that includes a processor and a memory having at least one instruction, at least one program, set of codes, or set of instructions stored therein. The at least one instruction, at least one program, set of codes, or set of instructions is configured to be executed by one or more processors to implement the above-described feature map generation method.

In an exemplary embodiment, a computer readable storage medium is also provided, in which at least one instruction, at least one program, a set of codes or a set of instructions is stored, which when executed by a processor of a computer device implements the above-mentioned characteristic data acquisition method.

In an exemplary embodiment, a computer readable storage medium is also provided, in which at least one instruction, at least one program, a set of codes or a set of instructions is stored, which when executed by a processor of a computer device implements the above data storage method.

In an exemplary embodiment, a computer readable storage medium is also provided, in which at least one instruction, at least one program, a set of codes, or a set of instructions is stored, which when executed by a processor of a computer device, implements the above-described feature map generation method.

Alternatively, the computer-readable storage medium may be a ROM (Read-Only Memory), a RAM (Random Access Memory), a CD-ROM (Compact Disc Read-Only Memory), a magnetic tape, a floppy disk, an optical data storage device, and the like.

In an exemplary embodiment, a computer program product is also provided, which, when executed, is adapted to carry out the above-mentioned characteristic data acquisition method.

In an exemplary embodiment, a computer program product is also provided for implementing the above-described data storage method when the computer program product is executed.

In an exemplary embodiment, a computer program product is also provided, which, when executed, is adapted to implement the above-described feature map generation method.

It should be understood that reference to "a plurality" herein means two or more. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. In addition, the step numbers described herein only exemplarily show one possible execution sequence among the steps, and in some other embodiments, the steps may also be executed out of the numbering sequence, for example, two steps with different numbers are executed simultaneously, or two steps with different numbers are executed in a reverse order to the order shown in the figure, which is not limited by the embodiment of the present application.

It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.

The above description is only exemplary of the present application and should not be taken as limiting the present application, and any modifications, equivalents, improvements and the like that are made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for feature data acquisition, the method comprising:

acquiring a characteristic map corresponding to a target service scene;

2. The method of claim 1, wherein the feature map comprises: the service feature, the service label, the attribute information of the service feature and the attribute information of the service label;

3. The method according to claim 2, wherein the obtaining feature data and tag data in the target service scenario according to the service feature and the service tag comprises:

and acquiring the label data which accords with the filtering rule of the label data.

4. The method of claim 2, wherein the service features include common service features and custom service features;

5. The method according to any one of claims 1 to 4, wherein after obtaining the feature data and the tag data in the target service scenario according to the service feature and the service tag, the method further comprises:

generating a training sample of a machine learning model according to the feature data and the label data;

and training the machine learning model by adopting the training samples, wherein the machine learning model is used for carrying out label classification on the target business scene.

6. The method of claim 5, wherein after training the machine learning model using the training samples, further comprising:

and/or the presence of a gas in the gas,

7. The method of claim 5, wherein after training the machine learning model using the training samples, further comprising:

generating an artificial intelligence AI template according to the machine learning model and the characteristic map;

and storing the AI template into a database of a service platform.

8. A method of data storage, the method comprising:

transmitting the feature map to a feature data generation device;

storing the machine learning model.

9. The method of claim 8, wherein after sending the feature map to a feature data generation device, further comprising:

receiving an Artificial Intelligence (AI) template sent by the feature data generation equipment, wherein the AI template is generated according to the machine learning model and the feature map;

and storing the AI template.

10. A method of feature pattern generation, the method comprising:

11. An apparatus for feature data acquisition, the apparatus comprising:

12. A data storage device, characterized in that the device comprises:

and the model storage module is used for storing the machine learning model.

13. A feature map generation apparatus, characterized in that the apparatus comprises:

14. A computer device comprising a processor and a memory, the memory having stored therein at least one instruction, at least one program, set of codes, or set of instructions, which is loaded and executed by the processor to implement the method of any one of claims 1 to 7, or to implement the method of any one of claims 8 to 9, or to implement the method of claim 10.

15. A computer readable storage medium having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions, which is loaded and executed by a processor to implement the method of any one of claims 1 to 7, or to implement the method of any one of claims 8 to 9, or to implement the method of claim 10.