TWM573022U - Management system for artificial intelligence knowledge - Google Patents

Abstract

The disclosure is related to a computer-implemented management system for artificial intelligence knowledge. The system includes an input management module used to manage the input data for performing artificial neural network upon an artificial intelligence model; an artificial intelligence model management module used to manage artificial intelligence models provided for selection; and an output management module used to manage the output data generated by performing the artificial neural network upon the artificial intelligence model. The system provides a calculation result management module used to manage the result in each calculation that is referred to adjusting the parameters incorporated into the artificial intelligence model for re-generating output data. A knowledge database is therefore constituted. A blockchain technology is adopted in the management system for creating distributed records over the multiple blockchain nodes.

Description

Artificial intelligence knowledge management system

The disclosure discloses a knowledge management system, in particular an artificial intelligence knowledge management system established by blockchain technology.

In the field of artificial intelligence (AI), it is one of the most important topics to establish an artificial intelligence model that can solve specific problems. For example, in some fields, it can be verified by repeated verification of successful artificial intelligence models such as human brain. Execution judgments such as image recognition, semantic analysis, games, etc.

The artificial neural network algorithm is one of the best tools for artificial model modeling. The neural network is composed of many nonlinear arithmetic units called neurons (Neuron), and these The numerous connections between neurons form a neural network that is usually operated in a parallel and decentralized manner, with the learning mechanism provided dependent on the activities of the neurons. In a neural network, there is a set of input neurons that are excited by specific data. After the excitation values are weighted and calculated by a function, the excitation values of the neurons are transmitted to other neurons. The process repeats repeatedly, stimulating the output neurons, and finally, the excitation value of this output neuron is the result of the calculation.

However, before reaching the expected results of the artificial intelligence model, the neural network calculus requires a large number of repeated calculus, trying multiple failures and adjustments before it is possible to make the output The results are close to expectations. However, the intermediate process does not establish a knowledge base that can be shared with the public, so that the latecomers who want to establish artificial intelligence still need to repeat past attempts and errors to get results.

In order to establish an artificial intelligence knowledge base that can be shared with the public, the latecomers can quickly develop in the footsteps of the predecessors, and maintain security and correctness. The specification discloses an artificial intelligence knowledge using blockchain technology. One of the main purposes of the management system is to safely and correctly share the process of developing an artificial intelligence model through blockchain technology, and become the knowledge base of later developers.

According to one embodiment, the artificial intelligence knowledge management system includes a computer system including a plurality of computer systems and software-implemented functional modules, including an input management module for managing a neural network used in developing an artificial intelligence model. Input data of the road algorithm; an artificial intelligence model management module for managing multiple artificial intelligence models to provide an artificial intelligence model; an output management module for managing the development of the artificial intelligence model with a neural network The output data generated by the algorithm; and a calculation result management module for managing data of the foregoing function module application, including adjusting parameters of the artificial intelligence model each time, and output data generated by the neural network algorithm each time .

In this way, further, the input data, the artificial intelligence model, the output data, and the parameters of each adjustment of the artificial intelligence model form a knowledge base, and the decentralized records dispersed in the plurality of blockchain nodes are formed by the blockchain technology. .

Further, the output data is compared with an expected value, and if not, the system provides a mechanism for adjusting the parameters of the artificial intelligence model.

Moreover, the artificial intelligence knowledge management system implements a cloud knowledge platform, and provides a blockchain record processing module to provide a query record function, so that the user of the artificial intelligence knowledge management system uses the computer device to query the blockchain technology. One of the artificial intelligence model's input data, output data and the parameters of each adjustment of the artificial intelligence model number.

In order to further understand the techniques, methods and effects of this creation in order to achieve the intended purpose, please refer to the following detailed descriptions and diagrams of this creation. I believe that the purpose, characteristics and characteristics of this creation can be deepened and specific. It is understood that the drawings are only for the purpose of illustration and description and are not intended to limit the invention.

10‧‧‧Artificial Intelligence Knowledge Management System

101‧‧‧Input Management Module

102‧‧‧Artificial Intelligence Model Management Module

103‧‧‧Output Management Module

104‧‧‧ calculus results management module

105‧‧‧ Knowledge Base

106‧‧‧ Blockchain Record Processing Module

107‧‧‧User Management Module

100‧‧‧Neuroid calculation module

20‧‧‧Network

201,202,203‧‧‧user node

21‧‧‧Artificial Intelligence Model Laboratory

22‧‧‧Artificial Intelligence Knowledge Management System

30‧‧ ‧ knowledge base

Step S301~S313‧‧‧ artificial intelligence knowledge management system operation process

Step S401~S413‧‧‧ Forming a process of decentralized records

Step S501~S507‧‧‧ artificial intelligence knowledge management system provides knowledge query process

1 is a schematic diagram showing an embodiment of an artificial intelligence knowledge management system architecture disclosed in the disclosure; FIG. 2 is a schematic diagram showing a relationship between an artificial intelligence knowledge management system and a participating user; FIG. 3 is a diagram showing an operational flow of the artificial intelligence knowledge management system; 4 shows an embodiment of a process in which an artificial intelligence knowledge management system forms a distributed record through a blockchain technology; and FIG. 5 shows an embodiment of a process in which an artificial intelligence knowledge management system provides a knowledge query.

Artificial Intelligence (AI) is a kind of computer science that realizes human intelligence through computer programs. It realizes reasoning, problem solving, learning, judgment, and even through computer programs such as the Artificial Neural Network algorithm. It is a step to replace decision-making. The advantage of using a computer program is that it can process large amounts of data and perform repetitive tasks, so that it can handle complex problems that humans are not good at. If artificial intelligence is to be able to operate correctly, or even replace part of human work, the correct artificial intelligence model is still needed. The way to build the model requires repeated and large amount of calculation, designing models and parameters, and finding algorithms to be correct. As a result, such as deep learning, which is a kind of machine learning, learns the correct direction from mistakes, and one of the ways to solve the problem is the neural network.

Today's neural networks are made up of many non-linear arithmetic units, which are generally called gods. The Neuron is composed of a number of links between these arithmetic units, and these arithmetic units are usually operated in a parallel and decentralized manner, so that a large amount of data can be processed at the same time, and such a design can be used. To handle a variety of applications that require a large amount of data operations, such as speech recognition, image recognition, and so on.

In the process of requiring huge data and processing volume, it really takes a lot of time cost. However, in the development and learning of many artificial intelligence, there is no smooth communication channel between the development teams, if there is convenient communication and learning. The pipeline will be shortened a lot of time. Thus, the specification discloses an artificial intelligence knowledge management system, which realizes AI knowledge management through the combination of software and hardware, and uses blockchain technology to record various AI knowledge in multiple On the decentralized nodes, the process of each team development can be effectively and correctly recorded, and all such knowledge can be conveniently queried in an environment with both rights management and security, and the time for artificial intelligence development can be shortened.

FIG. 1 is a schematic diagram showing an embodiment of an artificial intelligence knowledge management system architecture disclosed in the disclosure. This example shows an artificial intelligence knowledge management system 10 implemented by a computer system, which includes a functional module formed by combining hardware and software, and a computer system. The various software programs executed by one or more processors establish a cloud platform shared through the network, and use blockchain technology to spread the various AI development records obtained by the platform on the blockchain nodes.

The artificial intelligence model management module 102 manages a plurality of artificial intelligence models, and provides the platform user to select the required artificial intelligence model according to the authority, and the input management module 101 is used to develop an artificial intelligence model through software program management. Input data of a certain type of neural network algorithm (managed by the neural network calculus module 100); the output management module 103 is used to manage the output data generated by a certain type of neural network algorithm when developing the artificial intelligence model; The result management module 104 is configured to manage data applied by the input management module 101, the artificial intelligence model management module 102, and the output management module 103.

Since the neural network algorithm runs, according to the input data provided by the developer, it runs repeatedly according to the selected artificial intelligence model, and the output data is generated and developed. In the process, the output data can be compared with the expected value, and the result of the comparison becomes the basis for adjusting the parameters of the artificial intelligence model, and the calculation result management module 104 is used to manage the parameters of each adjustment of the artificial intelligence model, and each time the class is re-classified. The output data generated by the neural network algorithm.

The user in the system is managed by the user management module 107, and the user account and the authentication data of the artificial intelligence knowledge management system 10 are stored by a memory in the computer system. Moreover, the plurality of blockchain nodes may also include computer devices of users of the artificial intelligence knowledge management systems 10.

The input data, the artificial intelligence model, the output data, and the parameters of the artificial intelligence model are adjusted each time, or the expected value is added to form a knowledge base 105, and processed by the blockchain record processing module 106 in the artificial intelligence knowledge management system 10. The method includes processing each knowledge data into a packet distributed on the network, and forming a distributed record scattered on the plurality of blockchain nodes by the blockchain technology. The blockchain record processing module 106 is a core program of the system running blockchain, and each user of the artificial intelligence knowledge management system 10 establishes a blockchain account and obtains a key.

The blockchain record processing module 106 provides a query record function, allowing the user of the artificial intelligence knowledge management system 10 to use the computer device to connect to the artificial intelligence knowledge management system 10 through the blockchain record processing module 106. The blockchain technology queries the input data, the output data, and the parameters of the artificial intelligence model corresponding to one of the artificial intelligence models.

The neural network calculus module 100 represents each artificial intelligence development team, each of which runs an algorithm, for example, obtains various input data from the artificial intelligence knowledge management system 10, and generates various output data after calculation, including comparison expectations. As a result, the processes are all acquired and managed by the artificial intelligence knowledge management system 10. For example, the neural network-like algorithm may have multiple nodes (neurons) between the input layer and the output layer, such as through an artificial neural network (ANN), and each node has a different weight value. In order to become a complex function operation of artificial artificial thinking, the output value and the actual value obtained after the input value is simulated by the weight between nodes When there is a considerable drop in the inter-value, the system adjusts the parameters of the model and updates the weight operations between the nodes to bring the output values closer to the desired result.

It should be noted that the artificial intelligence knowledge management system 10 implements a cloud knowledge platform, and the neural network calculus module 100 can be an operation module of an external system, meaning that the artificial intelligence knowledge management system 10 proposed by the disclosure is a kind of The artificial intelligence knowledge management system can not participate in the development of artificial intelligence, but manage the platform of artificial intelligence knowledge through blockchain technology, so that developers of all parties can effectively obtain various kinds of security and query mechanisms provided by blockchain technology. Develop data and resources to improve development efficiency.

Figure 2 then shows a schematic diagram of the relationship between the artificial intelligence knowledge management system and the participating users.

In this example, the artificial intelligence model laboratory 21 is represented as a research and development unit that develops an artificial intelligence model. When developing an artificial intelligence model, it is necessary to determine various parameters in an artificial intelligence model and set input data. The neural network algorithm provides a mathematical model of AI deep learning for evaluation or approximation, and deep learning uses multiple layers. The neural network forms the input data and output data managed by the above system.

When running deep learning, take the image of an animal as an example, use the multi-layer neural network in the model, input a large number of animal pictures, let the computer program analyze the data by itself to find out the image feature value of the animal, let the computer learn as long as there is The higher the eigenvalue is, the more the image of the animal is connected. In the future, the computer will correctly recognize the image of the animal.

During the development of this artificial intelligence model, it is necessary to repeatedly check whether the output result is the result of correct recognition. If it does not, the parameters in the artificial intelligence model need to be re-adjusted. Thus, the artificial intelligence knowledge management system 22 obtains these inputs and outputs. Data such as calculation results form a knowledge base for artificial intelligence model development.

The artificial intelligence model laboratory 21 is connected to the artificial intelligence knowledge management system 22 by the network 20. The related experimental records further form a record scattered in each node by the blockchain technology, and the blockchain node can be implemented by various user nodes 201, 202, 203.

As another example, the artificial intelligence model can be a model for stock market forecasting analysis. When constructing this AI model, various parameters affecting stock market volatility become input data of the AI model, such as time (a certain year, month, day), stock number Various environmental variables, such as party support, gross national product, salary increase, climate, foreign investment ratio, etc., and the output data set is the expected price increase and decrease. The developer inputs the input data into the preset AI model, and the generated output data will be compared with the actual expected value. The error between the two becomes the basis for the developer to re-evaluate the parameters in the AI model, and the values in the input data can be adjusted. The weight, update the parameters in the model, if the input data is a one-dimensional matrix, the prediction is not allowed to turn into a multi-dimensional matrix, so that the output value is closer to the expected value.

The artificial intelligence knowledge management system 22 obtains the data generated during the process of establishing the AI model from the artificial intelligence model laboratory 21, and becomes an artificial intelligence knowledge base, which has also become a reference for other developers.

FIG. 3 depicts, by way of a flow, an operational flow embodiment of an artificial intelligence knowledge management system.

Beginning with step S301, in the development of the AI model, an artificial intelligence model is first introduced, and the related AI model can be obtained by the knowledge base 30. In step S303, relevant values are input according to the purpose of development, and these values are also one of the contents forming the knowledge base 30. Next, in step S305, when the AI model is developed, it is executed by the neural network algorithm, and the multi-layer neural network is calculated. In step S307, the result is outputted and becomes the content of the knowledge base 30.

Next, in step S309, the R&D team will evaluate whether it meets the requirements? If the expected output value is met (Yes), as in step S311, the AI model is confirmed, and the result also forms the content of the knowledge base 30; otherwise, if it does not meet the expectation (No), step S313 is executed to adjust the parameters in the AI model. The relevant parameters may also be part of the knowledge base 30 and recalculated.

Wherein, the input data, the artificial intelligence model, the output data, and the parameters of each adjustment of the artificial intelligence model form a knowledge base 30, and through the artificial intelligence knowledge management system, the blockchain technology can be used to form a plurality of blockchain nodes. For the decentralized recording of the system's user devices, these records scattered by blockchain technology are indispensable. Change, long-term record and convenient management and acquired characteristics.

An embodiment of a process for an artificial intelligence knowledge management system to form a decentralized record through a blockchain technique is depicted in FIG.

Initially, in step S401, the user connects the artificial intelligence knowledge management system through the computer device and forms a login user, as in step S403. At this time, the artificial intelligence knowledge management system assists the user to become the user and the node of the blockchain. In step S405, after the blockchain is activated, each user obtains the data sharing, encryption and decryption, and the key used for verification, and the manual is passed. The intelligent knowledge management system establishes a blockchain account number, as in step S407.

After that, the user of the artificial intelligence knowledge management system can obtain data through the blockchain, and may also participate in the research and development of the AI model, and the generated data establishes an artificial intelligence calculation record (step S409) to form a knowledge base in the artificial intelligence knowledge management system. Next, in step S411, the data is transferred to the block chain node, and in step S413, a distributed record is formed.

The artificial intelligence knowledge management system establishes a knowledge base for the development of the relevant AI model through the blockchain technology, and also provides a knowledge query process for the artificial knowledge management system shown in FIG. 5 for the knowledge base available for query.

In step S501, the user may first connect the artificial intelligence knowledge management system, such as step S503, after logging in to the system, authenticating the identity, and after confirming the permission according to step S505, the data may be queried through the specific user interface according to the authority. Specifically, as shown in step S507, the artificial intelligence calculation record is queried.

Thus, according to the above-mentioned artificial intelligence knowledge management system embodiment, the input data, the artificial intelligence model, the output data developed by the AI model and the parameters of each adjustment artificial intelligence model are formed into a knowledge base by computer technology, and further formed by blockchain technology. Decentralized records scattered across multiple blockchain nodes form an ecosystem of shared resources in the field of artificial intelligence, helping to shorten the artificial intelligence development timeline in related fields.

In order to further understand the techniques, methods and effects of this creation in order to achieve the intended purpose, please refer to the following detailed description and drawings of this creation. The purpose, characteristics, and characteristics of the creations are to be understood and understood in detail. However, the drawings are provided for reference and description only, and are not intended to limit the creation.

Claims (8)

An artificial intelligence knowledge management system comprising: a computer system comprising one or more processors, wherein: an input management module for managing input data of a type of neural network algorithm used in developing an artificial intelligence model; An artificial intelligence model management module for managing a plurality of artificial intelligence models to provide the artificial intelligence model; an output management module for managing output generated by the neural network algorithm when developing the artificial intelligence model Data; a calculation result management module for managing data input by the input management module, the artificial intelligence model management module, and the output management module, including adjusting parameters of the artificial intelligence model each time, and each time Retrieving the output data generated by the neural network algorithm; wherein the input data, the artificial intelligence model, the output data, and each parameter of the artificial intelligence model are formed into a knowledge base, and formed by a blockchain technique Decentralized records scattered across multiple blockchain nodes. The artificial intelligence knowledge management system of claim 1, wherein the parameters of the artificial intelligence model are adjusted according to the comparison of the output data with an expected value. The artificial intelligence knowledge management system of claim 2, wherein the expected value becomes one of data in the knowledge base that links the artificial intelligence model. The artificial intelligence knowledge management system of claim 1, further comprising a user management module for storing a user account and authentication data of the artificial intelligence knowledge management system through a memory in the computer system. The artificial intelligence knowledge management system of claim 4 further includes a blockchain record processing module for establishing a blockchain account for each user of the artificial intelligence knowledge management system and obtaining a key. The artificial intelligence knowledge management system of claim 5, wherein the plurality of blockchain nodes comprise a computer device of a user of the artificial intelligence knowledge management system. The artificial intelligence knowledge management system of claim 5, wherein the blockchain record processing module provides a query record function, so that the user of the artificial intelligence knowledge management system connects the network device through a network The artificial intelligence knowledge management system uses the blockchain recording processing module to query the input data, the output data, and the parameters of the artificial intelligence model corresponding to one of the artificial intelligence models by the blockchain technology. The artificial intelligence knowledge management system according to any one of claims 1 to 7, wherein the artificial intelligence knowledge management system implements a cloud knowledge platform.