CN111026858A

CN111026858A - Project information processing method and device based on project recommendation model

Info

Publication number: CN111026858A
Application number: CN201911204492.5A
Authority: CN
Inventors: 缪畅宇
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2019-11-29
Filing date: 2019-11-29
Publication date: 2020-04-17
Anticipated expiration: 2039-11-29
Also published as: CN111026858B

Abstract

The invention provides a project information processing method based on a project recommendation model, which comprises the following steps: acquiring item information in a use environment of the item recommendation model, and converting the item information into corresponding recognizable text information; generating text processing words corresponding to the word-level hidden variables and the selected probabilities of the text processing words through an encoder network of the project recommendation model and a decoder network of the project recommendation model according to the at least one word-level hidden variable and the corresponding fusion feature vector; selecting at least one text processing word to form a text processing result corresponding to the text information according to the selected probability of the text processing result; and converting the text processing result into new item information corresponding to the item recommendation model. The invention also provides a project information processing device, electronic equipment and a storage medium based on the project recommendation model. The invention can realize the matching of the new project information generated by the project recommendation model and the use environment.

Description

Project information processing method and device based on project recommendation model

Technical Field

The present invention relates to information processing technologies, and in particular, to a method and an apparatus for training a project recommendation model, an electronic device, and a storage medium.

Background

Human-Computer Interaction (HCI) refers to a process of information exchange between a person and a Computer determined in a certain interactive manner by using a certain dialogue language. With the development of human-computer interaction technology, more and more intelligent products based on human-computer interaction technology come into existence, for example, project recommendation can be completed through human-computer interaction in the project recommendation process, but in a traditional Seq2Seq model, projects in a generated sequence are high in popularity, high in occurrence frequency and too strong in universality, so that due to the limitation of RNN capacity, the project recommendation model is difficult to generate a high-quality text processing result, further the generation of project information is influenced, and the use experience of a user is also influenced.

Disclosure of Invention

In view of this, embodiments of the present invention provide a method, an apparatus, an electronic device, and a storage medium for processing item information based on an item recommendation model, and a technical solution of an embodiment of the present invention is implemented as follows:

the embodiment of the invention generally provides a project information processing method based on a project recommendation model, which comprises the following steps:

acquiring item information in a use environment of an item recommendation model, and converting the item information into corresponding identifiable text information;

determining at least one word-level hidden variable corresponding to text information through an encoder network of the project recommendation model;

generating, by a decoder network of the project recommendation model, text processing terms corresponding to the word-level hidden variables and a selected probability of the text processing terms according to the at least one word-level hidden variable and the corresponding fused feature vector;

selecting at least one text processing word to form a text processing result corresponding to the text information according to the selected probability of the text processing result;

and converting the text processing result into new item information corresponding to the item recommendation model so as to realize matching with the use environment.

The embodiment of the invention also provides a project information processing device based on the project recommendation model, which comprises:

the information transmission module is used for acquiring the project information in the using environment of the project recommendation model;

the information processing module is used for converting the project information into corresponding recognizable text information;

the information processing module is used for determining at least one word-level hidden variable corresponding to the text information through an encoder network of the project recommendation model;

the information processing module is used for generating text processing words corresponding to the hidden variables of the word level and the selected probability of the text processing words according to the hidden variables of the at least one word level and the corresponding fusion feature vectors through a decoder network of the project recommendation model;

the information processing module is used for selecting at least one text processing word to form a text processing result corresponding to the text information according to the selected probability of the text processing result;

and the information processing module is used for converting the text processing result into new item information corresponding to the item recommendation model so as to realize matching with the use environment.

In the above-mentioned scheme, the first step of the method,

the information processing module is used for determining a dynamic noise threshold value matched with the use environment of the item recommendation model;

the information processing module is used for converting the project information into initial text information;

the information processing module is used for carrying out denoising processing on the initial text information according to the dynamic noise threshold value and triggering a dynamic word segmentation strategy matched with the dynamic noise threshold value;

and the information processing module is used for performing word segmentation processing on the initial text information according to a dynamic word segmentation strategy matched with the dynamic noise threshold value to form corresponding identifiable text information.

In the above-mentioned scheme, the first step of the method,

the information processing module is used for determining a fixed noise threshold value matched with the use environment of the item recommendation model;

the information processing module is used for carrying out denoising processing on the initial text information according to the fixed noise threshold value and triggering a fixed word segmentation strategy matched with the fixed noise threshold value;

and the information processing module is used for performing word segmentation processing on the initial text information according to a fixed word segmentation strategy matched with the fixed noise threshold value to form corresponding identifiable text information.

In the above-mentioned scheme, the first step of the method,

the information processing module is used for determining a text processing result of the corresponding polling times according to the hidden variable of the at least one word level through a decoder network of the project recommendation model;

the information processing module is used for converting the text processing result of the corresponding polling times into a text processing result vector;

and the information processing module is used for generating text processing words corresponding to the word-level hidden variables and the selected probability of the text processing words according to the text processing result vector and the fusion feature vector through a decoder network of the project recommendation model.

In the above-mentioned scheme, the first step of the method,

the information processing module is used for determining a project diversity function matched with the project recommendation model;

the information processing module is used for adjusting the output result of the decoder network of the project recommendation model through the project diversity function so as to realize that the text processing words and the selected probability of the text processing words are matched with the project diversity function.

In the above-mentioned scheme, the first step of the method,

the information processing module is used for acquiring the user characteristics corresponding to the project recommendation model and forming a user characteristic vector according to the user characteristics;

the information processing module is used for acquiring the text processing result of the corresponding polling times and converting the text processing result of the corresponding polling times into a feature vector of a predicted text processing result;

and the information processing module is used for carrying out fusion processing on the user characteristic vector and the characteristic vector of the predicted text processing result to form a corresponding fusion characteristic vector.

In the above-mentioned scheme, the first step of the method,

the information processing module is used for, when the using environment of the item recommendation model is a video recommendation process,

and the information processing module is used for adjusting parameters of the cyclic convolution neural network based on the multiple attention mechanism in the decoder network according to the fusion feature vector of the project recommendation model so as to realize that the parameters of the cyclic convolution neural network based on the multiple attention mechanism are matched with the fusion feature vector.

In the above-mentioned scheme, the first step of the method,

the information processing module is used for acquiring a training sample matched with the use environment of the project recommendation model;

the information processing module is used for extracting a feature set matched with the training sample through the project recommendation model;

and the information processing module is used for training the project recommendation model according to the feature set matched with the training sample and the corresponding target text label so as to determine the model parameters matched with the project recommendation model.

In the above-mentioned scheme, the first step of the method,

the information processing module is used for sending the item information, the parameter information of the item recommendation model and new item information which is generated by the item recommendation model and is matched with the use environment to a blockchain network so as to enable the new item information to be matched with the use environment

And filling the project information, the parameter information of the project recommendation model and the new project information matched with the use environment into a new block by the node of the block chain network, and when the new block is identified in a consistent manner, adding the new block to the tail part of the block chain so as to realize that the project recommendation model can acquire the information in the block in different use environments of the same user.

An embodiment of the present invention further provides an electronic device, where the electronic device includes:

a memory for storing executable instructions;

and the processor is used for realizing the project information processing method based on the project recommendation model in the preamble when the executable instructions stored in the memory are run.

The embodiment of the invention also provides a computer-readable storage medium, which stores executable instructions and is characterized in that the executable instructions are executed by a processor to realize the project information processing method based on the preorder project recommendation model.

The embodiment of the invention has the following beneficial effects:

the method comprises the steps of obtaining item information in a use environment of an item recommendation model, and converting the item information into corresponding identifiable text information; determining at least one word-level hidden variable corresponding to text information through an encoder network of the project recommendation model; generating, by a decoder network of the project recommendation model, text processing terms corresponding to the word-level hidden variables and a selected probability of the text processing terms according to the at least one word-level hidden variable and the corresponding fused feature vector; selecting at least one text processing word to form a text processing result corresponding to the text information according to the selected probability of the text processing result; the text processing result is converted into new item information corresponding to the item recommendation model to be matched with the use environment, so that recommended items generated by the item recommendation model are more targeted, universal recommendation is reduced, user characteristics and the use environment matched with the recommendation model are better met, the richness and the foresight of the recommended items can be improved, and the use experience of users is improved.

Drawings

Fig. 1 is a schematic view of a usage scenario of a project information processing method of a project recommendation model according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a structure of an item information processing apparatus based on an item recommendation model according to an embodiment of the present invention;

FIG. 3 is a diagram illustrating the generation of item recommendations based on the RNN's Seq2Seq model in the prior art;

fig. 4 is an optional flowchart of a project information processing method of a project recommendation model according to an embodiment of the present invention;

FIG. 5 is an alternative flow chart of a project recommendation model training method according to an embodiment of the present invention;

FIG. 6 is a diagram illustrating an alternative process of an item recommendation model in an embodiment of the present invention;

FIG. 7 is a diagram illustrating an alternative process of the item recommendation method according to an embodiment of the present invention;

FIG. 8 is a block diagram of an architecture of a processing device 100 for an item recommendation model according to an embodiment of the present invention;

fig. 9 is a schematic structural diagram of a block chain in the block chain network 200 according to an embodiment of the present invention;

fig. 10 is a functional architecture diagram of a blockchain network 200 according to an embodiment of the present invention;

FIG. 11 is a diagram illustrating an application environment of a project recommendation model according to an embodiment of the present invention;

fig. 12 is an optional flowchart of a project information processing method based on a project recommendation model according to an embodiment of the present invention;

FIG. 13 is a schematic structural diagram of an item recommendation model according to an embodiment of the present invention;

FIG. 14 is a schematic view of a usage scenario of a project information processing method of a project recommendation model according to an embodiment of the present invention;

fig. 15 is a schematic view of a usage scenario of the item information processing method of the item recommendation model according to the embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail with reference to the accompanying drawings, the described embodiments should not be construed as limiting the present invention, and all other embodiments obtained by a person of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict.

Before further detailed description of the embodiments of the present invention, terms and expressions mentioned in the embodiments of the present invention are explained, and the terms and expressions mentioned in the embodiments of the present invention are applied to the following explanations.

1) In response to the condition or state on which the performed operation depends, one or more of the performed operations may be in real-time or may have a set delay when the dependent condition or state is satisfied; there is no restriction on the order of execution of the operations performed unless otherwise specified.

2) Word segmentation: also known as word segmentation, functions to segment the textual information of a complete sentence into a plurality of words, such as: liu De Hua is a Chinese singer. The result after word segmentation is: liu De Hua, China, singer.

3) A word bank is divided: the term segmentation library refers to a specific word segmentation method, and word dictionaries corresponding to different term segmentation libraries can be used for carrying out word segmentation processing on corresponding text information according to the word dictionaries corresponding to the term segmentation libraries.

4) Convolutional Neural Networks (CNN Convolutional Neural Networks) are a class of Feed forward Neural Networks (Feed forward Neural Networks) that contain convolution computations and have a deep structure, and are one of the representative algorithms for deep learning (deep). The convolutional neural network has a representation learning (representation learning) capability, and can perform shift-invariant classification (shift-invariant classification) on input information according to a hierarchical structure of the convolutional neural network.

5) And (4) model training, namely performing multi-classification learning on the image data set. The model can be constructed by adopting deep learning frames such as Tensor Flow, torch and the like, and a multi-classification model is formed by combining multiple layers of neural network layers such as CNN and the like. The input of the model is a three-channel or original channel matrix formed by reading an image through openCV and other tools, the output of the model is multi-classification probability, and the webpage category is finally output through softmax and other algorithms. During training, the model approaches to a correct trend through an objective function such as cross entropy and the like.

6) Neural Networks (NN): an Artificial Neural Network (ANN), referred to as Neural Network or Neural Network for short, is a mathematical model or computational model that imitates the structure and function of biological Neural Network (central nervous system of animals, especially brain) in the field of machine learning and cognitive science, and is used for estimating or approximating functions.

7) Machine Translation (MT): in the category of computational linguistics, the study of translating words or speech from one natural language to another by computer programs has been carried out. Neural Network Machine Translation (NMT) is a technique for performing Machine Translation using Neural network technology.

8) Encoder-decoder architecture: a network architecture commonly used for machine translation technology. The decoder receives the output result of the encoder as input and outputs a corresponding text sequence of another language.

9) Model parameters: is a number of functions that use generic variables to establish relationships between functions and variables. In artificial neural networks, the model parameters are typically real matrices.

10) API: the full-name Application Programming Interface can be understood as an Application program Interface semantically, and is some predefined functions or appointments for connection of different components of a software system. The goal is to provide applications and developers the ability to access a set of routines based on certain software or hardware without having to access native code or understand the details of the internal workings.

11) Transactions (transactions), equivalent to the computer term "Transaction," include operations that need to be committed to a blockchain network for execution and do not refer solely to transactions in the context of commerce, which embodiments of the present invention follow in view of the convention colloquially used in blockchain technology.

12) A Block chain (Blockchain) is a storage structure for encrypted, chained transactions formed from blocks (blocks).

13) A Blockchain Network (Blockchain Network) incorporates new blocks into a set of nodes of a Blockchain in a consensus manner.

14) Ledger (legger) is a general term for blockchains (also called Ledger data) and state databases synchronized with blockchains.

15) Intelligent Contracts (Smart Contracts), also known as chain codes (chaincodes) or application codes, are programs deployed in nodes of a blockchain network, and the nodes execute the intelligent Contracts called in received transactions to perform operations of updating or querying key-value data of a state database.

16) Consensus (Consensus), a process in a blockchain network, is used to agree on transactions in a block among a plurality of nodes involved, the agreed block is to be appended to the end of the blockchain, and the mechanisms for achieving Consensus include Proof of workload (PoW, Proof of Work), Proof of rights and interests (PoS, Proof of equity (DPoS), Proof of granted of shares (DPoS), Proof of Elapsed Time (PoET, Proof of Elapsed Time), and so on.

Fig. 1 is a schematic view of a usage scenario of a project information processing method of a project recommendation model according to an embodiment of the present invention, and referring to fig. 1, a client of project recommendation function software is disposed on a terminal (including a terminal 10-1 and a terminal 10-2), where in the present application, definitions of projects include, but are not limited to: music, videos, articles and commodities, such as a client or a plug-in for playing videos or a client with a shopping function, a user can obtain a target video through the corresponding client and can browse different commodity information; the terminal is connected to the server 200 through a network 300, and the network 300 may be a wide area network or a local area network, or a combination of the two, and uses a wireless link to realize data transmission.

As an example, the server 200 is configured to lay the item recommendation model and generate new item information matching the use environment through the item recommendation model, and display the new item information matching the use environment generated by the item recommendation model through the terminal (the terminal 10-1 and/or the terminal 10-2).

Specifically, the project information processing method of the project recommendation model comprises the following steps: acquiring item information in a use environment of an item recommendation model, and converting the item information into corresponding identifiable text information; determining at least one word-level hidden variable corresponding to text information through an encoder network of the project recommendation model; generating, by a decoder network of the project recommendation model, text processing terms corresponding to the word-level hidden variables and a selected probability of the text processing terms according to the at least one word-level hidden variable and the corresponding fused feature vector; selecting at least one text processing word to form a text processing result corresponding to the text information according to the selected probability of the text processing result; and converting the text processing result into new item information corresponding to the item recommendation model so as to realize matching with the use environment.

Of course, before the target question sentence is processed by the item recommendation model to generate corresponding new item information matched with the use environment, the item recommendation model needs to be trained, which specifically includes:

acquiring a training sample matched with the use environment of the project recommendation model; extracting a feature set matched with the training sample through the item recommendation model; and training the project recommendation model according to the feature set matched with the training sample and the corresponding target text label to determine model parameters matched with the project recommendation model. Further, the network parameters of the item recommendation model can be iteratively updated until corresponding convergence is reached, so that the training of the item recommendation model is realized.

To explain the structure of the project information processing apparatus based on the project recommendation model according to the embodiment of the present invention in detail, the project information processing apparatus based on the project recommendation model may be implemented in various forms, such as a dedicated terminal with a project recommendation model processing function, or a server provided with a project recommendation model processing function, such as the server 200 in the foregoing fig. 1. Fig. 2 is a schematic diagram of a configuration of an item information processing apparatus based on an item recommendation model according to an embodiment of the present invention, and it is understood that fig. 2 only shows an exemplary configuration of the item information processing apparatus based on the item recommendation model, and not a whole configuration, and a part of the configuration or the whole configuration shown in fig. 2 may be implemented as needed.

The project information processing device based on the project recommendation model provided by the embodiment of the invention comprises: at least one processor 201, memory 202, user interface 203, and at least one network interface 204. The various components in the project information processing apparatus based on the project recommendation model are coupled together by a bus system 205. It will be appreciated that the bus system 205 is used to enable communications among the components. The bus system 205 includes a power bus, a control bus, and a status signal bus in addition to a data bus. For clarity of illustration, however, the various buses are labeled as bus system 205 in fig. 2.

The user interface 203 may include, among other things, a display, a keyboard, a mouse, a trackball, a click wheel, a key, a button, a touch pad, or a touch screen.

It will be appreciated that the memory 202 can be either volatile memory or nonvolatile memory, and can include both volatile and nonvolatile memory. The memory 202 in embodiments of the present invention is capable of storing data to support operation of the terminal (e.g., 10-1). Examples of such data include: any computer program, such as an operating system and application programs, for operating on a terminal (e.g., 10-1). The operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, and the like, and is used for implementing various basic services and processing hardware-based tasks. The application program may include various application programs.

In some embodiments, the item information processing apparatus based on the item recommendation model according to the embodiments of the present invention may be implemented by combining software and hardware, and as an example, the item information processing method and apparatus based on the item recommendation model according to the embodiments of the present invention may be a processor in the form of a hardware decoding processor, which is programmed to execute the item information processing method based on the item recommendation model according to the embodiments of the present invention. For example, a processor in the form of a hardware decoding processor may employ one or more Application Specific Integrated Circuits (ASICs), DSPs, Programmable Logic Devices (PLDs), Complex Programmable Logic Devices (CPLDs), Field Programmable Gate Arrays (FPGAs), or other electronic components.

As an example of the implementation of the item information processing apparatus based on the item recommendation model according to the embodiment of the present invention by combining software and hardware, the item information processing apparatus based on the item recommendation model according to the embodiment of the present invention may be directly embodied as a combination of software modules executed by the processor 201, the software modules may be located in a storage medium, the storage medium is located in the memory 202, and the processor 201 reads executable instructions included in the software modules in the memory 202 and completes the item information processing method of the item recommendation model according to the embodiment of the present invention by combining necessary hardware (for example, including the processor 201 and other components connected to the bus 205).

By way of example, the Processor 201 may be an integrated circuit chip having Signal processing capabilities, such as a general purpose Processor, a Digital Signal Processor (DSP), or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or the like, wherein the general purpose Processor may be a microprocessor or any conventional Processor or the like.

As an example of the hardware implementation of the project information processing apparatus based on the project recommendation model provided in the embodiment of the present invention, the apparatus provided in the embodiment of the present invention may be implemented directly by using the processor 201 in the form of a hardware decoding processor, for example, the project information processing method for implementing the project recommendation model provided in the embodiment of the present invention may be implemented by one or more Application Specific Integrated Circuits (ASICs), DSPs, Programmable Logic Devices (PLDs), Complex Programmable Logic Devices (CPLDs), Field Programmable Gate Arrays (FPGAs), or other electronic components.

The memory 202 in the embodiment of the present invention is used to store various types of data to support the operation of the item information processing apparatus based on the item recommendation model. Examples of such data include: any executable instructions for operating on an item information processing apparatus based on an item recommendation model, such as executable instructions, a program that implements the item information processing method from the item recommendation model of the embodiment of the present invention may be contained in the executable instructions.

In other embodiments, the item information processing apparatus based on the item recommendation model according to the embodiments of the present invention may be implemented in software, and fig. 2 shows the item information processing apparatus based on the item recommendation model stored in the memory 202, which may be software in the form of programs, plug-ins, and the like, and includes a series of modules, and as an example of the programs stored in the memory 202, the item information processing apparatus based on the item recommendation model may include the following software modules: an information transmission module 2081 and an information processing module 2082. When the software modules in the project information processing apparatus based on the project recommendation model are read into the RAM by the processor 201 and executed, the project information processing method of the project recommendation model according to the embodiment of the present invention will be implemented, and the following will continue to describe the functions of the software modules in the project information processing apparatus based on the project recommendation model, wherein,

the information transmission module 2081 is used for acquiring the item information in the use environment of the item recommendation model.

The information processing module 2082 is used for converting the item information into corresponding recognizable text information;

the information processing module 2082 is configured to determine at least one word-level hidden variable corresponding to text information through the encoder network of the item recommendation model;

the information processing module 2082 is configured to generate, through the decoder network of the project recommendation model, a text processing word corresponding to the word-level hidden variable and a selected probability of the text processing word according to the at least one word-level hidden variable and the corresponding fusion feature vector;

the information processing module 2082 is configured to select at least one text processing word to form a text processing result corresponding to the text information according to the selected probability of the text processing result;

the information processing module 2082 is configured to convert the text processing result into new item information corresponding to the item recommendation model, so as to implement matching with the usage environment.

Before describing the method for processing item information based on an item recommendation model provided by the embodiment of the present invention, first, in a process of generating a corresponding reply sentence according to a question text by a conventional item recommendation model in the related art, before describing the method for processing item information based on an item recommendation model provided by the embodiment of the present invention, fig. 3 is a schematic diagram of generating an item recommendation by a Seq2Seq model based on an RNN in the prior art, and fig. 2 is a schematic diagram of operating a conventional item recommendation model, wherein in an operating process of an item recommendation model shown in fig. 2, a Seq2Seq model based on an RNN generates a recommendation item, wherein eq2Seq model is an architectural manner represented by an encoder (Encode) and a decoder (Decode), and Seq2Seq model generates an output sequence Y according to an input sequence X. In the seq2seq model represented by an encoder (Encode) which converts an input sequence into a vector of fixed length, and a decoder (Decode) which decodes the input vector of fixed length into an output sequence. As shown in fig. 2, an Encoder (Encoder) encodes input text information to obtain corresponding text features; the Decoder (Decoder) decodes the text features and outputs the decoded text features to generate corresponding text processing results (i.e. new item information), wherein the encoder (Encode) and the Decoder (Decode) are in one-to-one correspondence.

It can be seen that, for the related art shown in fig. 2, the disadvantage of the project recommendation model based on the Seq2Seq model is that the model in the related art only establishes a one-to-one relationship for the training data Query-Response, and uses MLE to optimize the model, which results in that the model generates many high-frequency general projects, so that the projects in the generated sequence have high heat, high frequency and strong universality, and therefore, due to the limitation of RNN capability, the project recommendation model is difficult to generate high-quality text processing results, which further affects the generation of project information and also affects the user experience

To solve the defects in the related art, referring to fig. 4, fig. 4 is an optional flowchart of a project information processing method of a project recommendation model according to an embodiment of the present invention, and it can be understood that the steps shown in fig. 4 may be executed by various electronic devices operating the project recommendation model project information processing method apparatus, for example, a dedicated terminal with a model training function, a server with a project recommendation model training function, or a server cluster. The following is a description of the steps shown in fig. 4.

Step 401: the item information processing apparatus acquires item information in a use environment of an item recommendation model and converts the item information into corresponding recognizable text information.

In some embodiments of the present invention, obtaining item information in an environment where an item recommendation model is used, and converting the item information into corresponding recognizable text information, comprises:

determining a dynamic noise threshold value matched with the use environment of the item recommendation model; converting the project information into initial text information; denoising the initial text information according to the dynamic noise threshold value, and triggering a dynamic word segmentation strategy matched with the dynamic noise threshold value; and performing word segmentation processing on the initial text information according to a dynamic word segmentation strategy matched with the dynamic noise threshold value to form corresponding identifiable text information. For example, in the use environment of academic translation (sorting and outputting the generated machine translation results), the dynamic noise threshold value of the item information displayed by the terminal, which only includes the text information of the academic paper and matches with the use environment of the item recommendation model, needs to be smaller than that in the use environment of the social news information translation.

The method comprises the following steps that item information is required to be converted into corresponding recognizable text information for processing by a recommendation model no matter video recommendation, commodity recommendation or text recommendation (machine translation or voice recognition), so-called word segmentation, namely verb meaning and name meaning, is achieved; each participle is a word or a phrase, namely the minimum semantic unit with definite meaning; for the received use environments of different users or different text processing models, the minimum semantic units contained in the received use environments need to be divided into different types, and adjustment needs to be made timely, and the process is called word segmentation, namely the word segmentation can refer to the process for dividing the minimum semantic units; on the other hand, the minimum semantic unit obtained after division is also often called word segmentation, that is, a word obtained after the word segmentation is performed; in order to distinguish the two meanings from each other, the smallest semantic unit referred to by the latter meaning is sometimes referred to as a participle object (Term); the term participled object is used in this application; the word segmentation object corresponds to a keyword which is used as an index basis in the inverted list. For Chinese, because words as the minimum semantic unit are often composed of different numbers of characters, and there are no natural distinguishing marks in alphabetic writing such as blank partitions and the like between the words, it is an important step for Chinese to accurately perform word segmentation to obtain reasonable word segmentation objects.

In some embodiments of the present invention, the language habits and the operation habits of different users are different, and different word segmentation methods need to be adjusted for different users to adapt to the language habits of different users. Especially for Chinese, the meaning unit is expressed based on Chinese characters, and the minimum semantic unit which really has a meaning is a word; because the space between words is not used as the segmentation like the space between English words, which words form words in a sentence of text is uncertain, and therefore, the word segmentation of Chinese texts is an important work. Moreover, for the text processing instruction text which contains some things which are only valuable for natural language understanding, and for the text processing model, to inquire related contents, it is necessary to determine which are really valuable retrieval bases, so that through denoising processing, a word-level feature vector set corresponding to the text processing instruction text can be formed, and the occurrence of meaningless word-level feature vectors such as 'of', 'ground' and 'get' in the word-level feature vector set is avoided "

determining a fixed noise threshold that matches the usage environment of the item recommendation model; converting the project information into initial text information; denoising the initial text information according to the fixed noise threshold value, and triggering a fixed word segmentation strategy matched with the fixed noise threshold value; and performing word segmentation processing on the initial text information according to a fixed word segmentation strategy matched with the fixed noise threshold value to form corresponding identifiable text information. When the project recommendation model is solidified in a corresponding hardware mechanism, such as a vehicle-mounted terminal or an intelligent medical system, and the using environment is a professional term question sentence (or a question sentence in a certain field), because the noise is relatively single, the processing speed of the project recommendation model can be effectively increased, the waiting time of a user is reduced, and the using experience of the user is improved through a fixed noise threshold corresponding to the fixed project recommendation model.

Step 402: the project information processing device determines at least one word-level hidden variable corresponding to the text information through an encoder network of the project recommendation model.

Step 403: and the project information processing device generates text processing words corresponding to the hidden variables of the word level and the selected probability of the text processing words according to the hidden variables of the at least one word level and the corresponding fusion feature vectors through a decoder network of the project recommendation model.

In some embodiments of the invention, the method further comprises:

acquiring user characteristics corresponding to the project recommendation model, and forming a user characteristic vector according to the user characteristics; acquiring a text processing result of corresponding polling times, and converting the text processing result of the corresponding polling times into a feature vector of a predicted text processing result; and fusing the user feature vector and the feature vector of the predicted text processing result to form a corresponding fused feature vector. Because different users have different preferences for the output result of the item recommendation model, the traditional item recommendation model shown in fig. 3 often ignores the influence of the user on the item recommendation model, and therefore, the influence of the characteristics of different users on the model output result can be fully considered when the item recommendation model forms new item information by forming a user characteristic vector according to the user characteristics. Further, the text processing result of the corresponding polling times is converted into the feature vector of the predicted text processing result, so that the project recommendation model can fully consider the influence of the predicted text processing result on the amplitude of the output result in the iterative processing of the decoder, and the amplitude of the decoder network output result of the project recommendation model is expanded.

In some embodiments of the present invention, the generating, by the decoder network of the item recommendation model, the text processing words corresponding to the word-level hidden variables and the selected probabilities of the text processing words according to the at least one word-level hidden variables and the corresponding fused feature vectors may be implemented by:

determining a text processing result of the corresponding polling times according to the hidden variable of the at least one word level through a decoder network of the project recommendation model; converting the text processing result of the corresponding polling times into a text processing result vector; and generating text processing words corresponding to the hidden variables of the word level and the selected probability of the text processing words according to the text processing result vector and the fusion feature vector through a decoder network of the project recommendation model.

In some embodiments of the invention, the method further comprises:

determining an item diversity function matching the item recommendation model; and adjusting the output result of the decoder network of the project recommendation model through the project diversity function to realize that the text processing words and the selected probability of the text processing words are matched with the project diversity function.

Step 404: and the project information processing device selects at least one text processing word to form a text processing result corresponding to the text information according to the selection probability of the text processing result.

Step 405: and the item information processing device converts the text processing result into new item information corresponding to the item recommendation model so as to realize matching with the use environment.

In some embodiments of the invention, when the usage environment of the item recommendation model is a video recommendation process,

and adjusting parameters of a cyclic convolution neural network based on the multiple attention mechanism in the decoder network according to the fusion feature vector of the project recommendation model so as to realize that the parameters of the cyclic convolution neural network based on the multiple attention mechanism are matched with the fusion feature vector. The fused feature vector can be processed through a cyclic convolution neural network based on a multiple attention mechanism in the decoder network, so that the influence of the predicted text processing result on the amplitude of the output result in the iterative processing is fully considered, and the amplitude of the decoder network output result of the project recommendation model is expanded.

With continuing reference to fig. 5, fig. 5 is an optional flowchart of the project recommendation model training method according to the embodiment of the present invention, and it can be understood that the steps shown in fig. 5 may be executed by various electronic devices of the question and sentence processing apparatus running an answer model, for example, a dedicated terminal with a project recommendation model training function, a server with a project recommendation model training function, or a server cluster. The following is a description of the steps shown in fig. 5.

Step 501: the project information processing device acquires a training sample matched with the use environment of the project recommendation model.

Step 502: and the project information processing device extracts a characteristic set matched with the training sample through the project recommendation model.

Step 503: and the project information processing device trains the project recommendation model according to the feature set matched with the training sample and the corresponding target text label so as to determine model parameters matched with the project recommendation model.

In some embodiments of the present invention, the processing of the training sample set by the item recommendation model to determine the updated parameters of the item recommendation model in response to the initial parameters of the item recommendation model may be implemented by:

substituting different statement samples in the training sample set into a loss function corresponding to a network structure consisting of an encoder network and a decoder network of the project recommendation model; and determining network parameters of an encoder network and a decoder network corresponding to the item recommendation model when the loss function meets the corresponding convergence condition as update parameters of the item recommendation model.

In some embodiments of the present invention, the network parameters of the item recommendation model may be iteratively updated through the corresponding training sample set according to the updated parameters of the item recommendation model until corresponding convergence is reached.

Wherein, the loss function of the item recommendation model is expressed as:

loss _ a ═ Σ (decoder _ a (encoder (warp (x1))) -x1) 2; where decoder _ a is decoder a, warp is a function of the question statement used for training, x1 is the question statement, and encoder is the encoder.

In the iterative training process, by substituting the problem statement into the loss function of the project recommendation model, parameters of the encoder A and the decoder A when the loss function is reduced according to the gradient (such as the maximum gradient) are solved, and when the loss function is converged (namely, when the hidden variable capable of forming the word level corresponding to the problem statement is determined), the training is finished.

In the training process of the item recommendation model, the loss function of the item recommendation model is represented as: loss _ B ═ Σ (decoder _ B (encoder (warp (x2))) -x2) 2; where decoder _ B is the decoder B, warp is the function of the question statement used for training, x2 is the question statement, and encoder is the encoder.

In the iterative training process, solving parameters of an encoder B and a decoder B when a loss function descends according to a gradient (such as a maximum gradient) by substituting a problem statement into a loss function of a project recommendation model; when the loss function converges (i.e., when the decoding results in a selected probability of a reply sentence corresponding to the question sentence), the adaptation and training ends.

Referring to fig. 6, fig. 6 is a schematic diagram of an optional processing procedure of the project recommendation model in the embodiment of the present invention, where the encoder may include a convolutional neural network, and after inputting the feature vector set into the encoder, outputs a corresponding floating point feature vector corresponding to the feature vector set. Specifically, the feature vector set is input into the encoder, that is, the convolutional neural network in the encoder, the corresponding floating point feature vector corresponding to the feature vector set is extracted through the convolutional neural network, the convolutional neural network outputs the extracted corresponding floating point feature vector and serves as the output of the encoder, and then the phase execution is performed by using the floating point feature vector output by the encoderThe corresponding text information processing, or the encoder may include a convolutional neural network and a cyclic neural network, and after the feature vector set is input to the encoder, the corresponding floating point feature vector carrying the timing information corresponding to the feature vector set is output, as shown in the encoder in fig. 6. Specifically, the feature vector set is input to the encoder, i.e., a convolutional neural network (e.g., CNN neural network in fig. 6) in the encoder, a corresponding floating point feature vector corresponding to the feature vector set is extracted by the convolutional neural network, the convolutional neural network outputs the extracted corresponding floating point feature vector, and the cyclic neural network (corresponding to h in fig. 6) in the encoder is input_i-1、h_iAnd the like) to extract and fuse the timing information of the extracted convolutional neural network characteristic vector through a recurrent neural network, wherein the recurrent neural network outputs a floating point characteristic vector carrying the timing information and serves as the output of an encoder, and then corresponding processing steps are executed by using the floating point characteristic vector output by the encoder.

Referring to fig. 7, fig. 7 is a schematic diagram of an optional processing procedure of the project recommendation method according to the embodiment of the present invention, where the dual-flow long-short term memory network may include a bidirectional vector model, an attention model, a full-link layer, and a sigmoid classifier, the bidirectional vector model performs recursion processing on different feature vectors in a feature vector set of an input text to be processed, and combines the feature vectors after the recursion processing together to form a longer vector, for example, combining part-of-speech feature vectors together to form a longer vector, and combines the two combined vectors together again to form a longer vector, and finally uses the two full-link layers to map learned distributed feature representations to corresponding sample label spaces to improve accuracy of a final classification result, and finally uses the sigmoid classifier to determine probability values of the text to be processed corresponding to respective labels, and integrating the item recommendation result to form new text information corresponding to the text information.

In some embodiments of the invention, the method further comprises:

sending the item information, the parameter information of the item recommendation model and new item information which is generated by the item recommendation model and is matched with the use environment to a blockchain network so as to enable the item information, the parameter information of the item recommendation model and the new item information to be matched with the use environment in the use environment

Referring to fig. 8, fig. 8 is a schematic architecture diagram of a processing apparatus 100 of an item recommendation model according to an embodiment of the present invention, which includes a blockchain network 200 (exemplarily illustrating a consensus node 210-1 to a consensus node 210-3), an authentication center 300, a service agent 400, and a service agent 500, which are respectively described below.

The type of blockchain network 200 is flexible and may be, for example, any of a public chain, a private chain, or a federation chain. Taking a public link as an example, electronic devices such as user terminals and servers of any service entity can access the blockchain network 200 without authorization; taking a federation chain as an example, an electronic device (e.g., a terminal/server) under the jurisdiction of a service entity after obtaining authorization may access the blockchain network 200, and at this time, become a client node in the blockchain network 200.

In some embodiments, the client node may act as a mere watcher of the blockchain network 200, i.e., provides functionality to support a business entity to initiate a transaction (e.g., for uplink storage of data or querying of data on a chain), and may be implemented by default or selectively (e.g., depending on the specific business requirements of the business entity) with respect to the functions of the consensus node 210 of the blockchain network 200, such as a ranking function, a consensus service, and an accounting function, etc. Therefore, the data and the service processing logic of the service subject can be migrated into the block chain network 200 to the maximum extent, and the credibility and traceability of the data and service processing process are realized through the block chain network 200.

Consensus nodes in blockchain network 200 receive transactions submitted from client nodes (e.g., client node 410 shown in fig. 8 as belonging to business entity 400 and client node 510 shown in fig. 8 as belonging to server 500) of different business entities (e.g., business entity 400 and business entity 500 shown in fig. 8), perform the transactions to update the ledger or query the ledger, and various intermediate or final results of performing the transactions may be returned for display in the client nodes of the business entities.

For example, the client node 410/510 may subscribe to events of interest in the blockchain network 200, such as transactions occurring in a particular organization/channel in the blockchain network 200, and the corresponding transaction notifications are pushed by the consensus node 210 to the client node 410/510, thereby triggering the corresponding business logic in the client node 410/510.

An exemplary application of the blockchain network is described below, taking an example in which a plurality of service entities access the blockchain network to manage and process project information.

Referring to fig. 8, a plurality of business entities involved in the management link, such as the business entity 400, may be processing devices based on an artificial intelligence project recommendation model, the business entity 500 may be a display system with a text display (operation) function, and registers from the certificate authority 300 to obtain respective digital certificates, each digital certificate includes a public key of the business entity and a digital signature signed by the certificate authority 300 on the public key and identity information of the business entity, and is used for being attached to a transaction together with the digital signature of the business entity on the transaction and sent to a blockchain network, for the blockchain network to take out the digital certificate and signature from the transaction, verify the authenticity of the message (i.e. whether it has not been tampered with) and the identity information of the service entity sending the message, and the blockchain network will verify according to the identity, for example whether it has the right to initiate the transaction. Clients running on electronic devices (e.g., terminals or servers) hosted by the business entity may request access from the blockchain network 200 to become client nodes.

The client node 410 of the service body 400 is used for acquiring the item information in the use environment of the item recommendation model and converting the item information into corresponding recognizable text information;

determining at least one word-level hidden variable corresponding to text information through an encoder network of the project recommendation model; generating, by a decoder network of the project recommendation model, text processing terms corresponding to the word-level hidden variables and a selected probability of the text processing terms according to the at least one word-level hidden variable and the corresponding fused feature vector; selecting at least one text processing word to form a text processing result corresponding to the text information according to the selected probability of the text processing result; and converting the text processing result into new item information corresponding to the item recommendation model so as to match the use environment, and sending the item information, the parameter information of the item recommendation model and the new item information which is generated by the item recommendation model and is matched with the use environment to the blockchain network 200.

The item information, the parameter information of the item recommendation model, and the new item information generated by the item recommendation model and matching with the use environment are sent to the blockchain network 200, service logic may be set in the client node 410 in advance, when corresponding item information is formed, the client node 410 automatically sends the new item information matching with the use environment to the blockchain network 200, or a service person of the service agent 400 logs in the client node 410, manually packages the new item information matching with the use environment, and sends the new item information to the blockchain network 200. Upon transmission, the client node 410 generates a transaction corresponding to the update operation based on the new item information that matches the usage environment, specifies in the transaction the smart contract that needs to be invoked to implement the update operation, and the parameters passed to the smart contract, and the transaction also carries the digital certificate of the client node 410, a signed digital signature (e.g., encrypted using a private key in the digital certificate of the client node 410 for a digest of the transaction), and broadcasts the transaction to the consensus node 210 in the blockchain network 200.

When the transaction is received in the consensus node 210 in the blockchain network 200, the digital certificate and the digital signature carried by the transaction are verified, after the verification is successful, whether the service agent 400 has the transaction right is determined according to the identity of the service agent 400 carried in the transaction, and the transaction fails due to any verification judgment of the digital signature and the right verification. After successful verification, node 210 signs its own digital signature (e.g., by encrypting the digest of the transaction using the private key of node 210-1) and continues to broadcast in blockchain network 200.

After receiving the transaction successfully verified, the consensus node 210 in the blockchain network 200 fills the transaction into a new block and broadcasts the new block. When a new block is broadcasted by the consensus node 210 in the block chain network 200, performing a consensus process on the new block, if the consensus is successful, adding the new block to the tail of the block chain stored in the new block, updating the state database according to a transaction result, and executing a transaction in the new block: adding, in a status database, a key-value pair including item information, parameter information of the item recommendation model, and new item information matching in the use environment for a transaction submitting updated item information, parameter information of the item recommendation model, and new item information matching in the use environment.

A service person of the service agent 500 logs in the client node 510, inputs item information or an item information query request, the client node 510 generates a transaction corresponding to an update operation/query operation according to the item information or the item information query request, specifies an intelligent contract that needs to be called to implement the update operation/query operation and parameters transferred to the intelligent contract in the transaction, and broadcasts the transaction to the consensus node 210 in the blockchain network 200, where the transaction also carries a digital certificate of the client node 510 and a signed digital signature (for example, a digest of the transaction is encrypted by using a private key in the digital certificate of the client node 510).

After receiving the transaction in the consensus node 210 in the blockchain network 200, verifying the transaction, filling the block and making the consensus consistent, adding the filled new block to the tail of the blockchain stored in the new block, updating the state database according to the transaction result, and executing the transaction in the new block: for the submitted transaction of updating the manual identification result corresponding to certain item information, updating the key value pair corresponding to the item information in the state database according to the manual identification result; and for the submitted transaction for inquiring certain item information, inquiring the key value pair corresponding to the item information from the state database, and returning a transaction result.

It should be noted that fig. 8 illustrates a process of directly linking the project information, the parameter information of the project recommendation model, and the new project information matching the usage environment, but in other embodiments, for a large data volume of the project information, the client node 410 may pair and link the hash of the project information and the corresponding hash of the project information, and store the original project information and the corresponding project information in a distributed file system or a database. After the client node 510 obtains the item information and the corresponding item information from the distributed file system or the database, it may perform a check in combination with the corresponding hash in the blockchain network 200, thereby reducing the workload of the uplink operation.

As an example of a block chain, referring to fig. 9, fig. 9 is a schematic structural diagram of a block chain in a block chain network 200 according to an embodiment of the present invention, where a header of each block may include hash values of all transactions in the block and also include hash values of all transactions in a previous block, a record of a newly generated transaction is filled in the block and is added to a tail of the block chain after being identified by nodes in the block chain network, so as to form a chain growth, and a chain structure based on hash values between blocks ensures tamper resistance and forgery prevention of transactions in the block. The project information stored in the blockchain network can be a dedicated text in a certain field (for example, case information of a medical system or an experimental information data text in a scientific experiment), and the project information can be shared among different nodes by storing the project information in the blockchain network.

An exemplary functional architecture of a block chain network provided in the embodiment of the present invention is described below, referring to fig. 10, fig. 10 is a schematic functional architecture diagram of a block chain network 200 provided in the embodiment of the present invention, which includes an application layer 201, a consensus layer 202, a network layer 203, a data layer 204, and a resource layer 205, which are described below respectively.

The resource layer 205 encapsulates the computing, storage, and communication resources that implement each node 210 in the blockchain network 200.

The data layer 204 encapsulates various data structures that implement the ledger, including blockchains implemented in files in a file system, state databases of the key-value type, and presence certificates (e.g., hash trees of transactions in blocks).

The network layer 203 encapsulates the functions of a Point-to-Point (P2P) network protocol, a data propagation mechanism and a data verification mechanism, an access authentication mechanism and service agent identity management.

Wherein the P2P network protocol implements communication between nodes 210 in the blockchain network 200, the data propagation mechanism ensures propagation of transactions in the blockchain network 200, and the data verification mechanism implements reliability of data transmission between nodes 210 based on cryptography methods (e.g., digital certificates, digital signatures, public/private key pairs); the access authentication mechanism is used for authenticating the identity of the service subject added into the block chain network 200 according to an actual service scene, and endowing the service subject with the authority of accessing the block chain network 200 when the authentication is passed; the business entity identity management is used to store the identity of the business entity that is allowed to access blockchain network 200, as well as the permissions (e.g., the types of transactions that can be initiated).

The consensus layer 202 encapsulates the functions of the mechanism for the nodes 210 in the blockchain network 200 to agree on a block (i.e., a consensus mechanism), transaction management, and ledger management. The consensus mechanism comprises consensus algorithms such as POS, POW and DPOS, and the pluggable consensus algorithm is supported.

The transaction management is configured to verify a digital signature carried in the transaction received by the node 210, verify identity information of the service entity, and determine whether the node has an authority to perform the transaction (read related information from the identity management of the service entity) according to the identity information; for the service agents authorized to access the blockchain network 200, the service agents all have digital certificates issued by the certificate authority, and the service agents sign the submitted transactions by using private keys in the digital certificates of the service agents, so that the legal identities of the service agents are declared.

The ledger administration is used to maintain blockchains and state databases. For the block with the consensus, adding the block to the tail of the block chain; executing the transaction in the acquired consensus block, updating the key-value pairs in the state database when the transaction comprises an update operation, querying the key-value pairs in the state database when the transaction comprises a query operation and returning a query result to the client node of the business entity. Supporting query operations for multiple dimensions of a state database, comprising: querying the block based on the block vector number (e.g., hash value of the transaction); inquiring the block according to the block hash value; inquiring a block according to the transaction vector number; inquiring the transaction according to the transaction vector number; inquiring account data of a business main body according to an account (vector number) of the business main body; and inquiring the block chain in the channel according to the channel name.

The application layer 201 encapsulates various services that the blockchain network can implement, including tracing, crediting, and verifying transactions.

The following describes a project information processing method based on a project recommendation model according to an embodiment of the present invention with an in-vehicle usage environment, where fig. 11 is a schematic view of an application environment of the project recommendation model according to the embodiment of the present invention, where, referring to fig. 11, a client capable of displaying software of a corresponding project is disposed on an in-vehicle terminal (including a terminal 110-1 and a terminal 110-2), where in the present application, definitions of the project include, but are not limited to: the system comprises a vehicle-mounted system, a client, a server and a client side, wherein the vehicle-mounted system comprises a music, video, articles and commodities, for example, a video playing client or a plug-in, or a shopping function client, and a user can obtain a target video through a corresponding client, browse different commodity information or obtain corresponding services in a vehicle environment; the terminal is connected to the server 200 through a network 300, and the network 300 may be a wide area network or a local area network, or a combination of the two, and uses a wireless link to realize data transmission. In the process, in order to recommend more accurate item information to the user, a corresponding recommendation system can be used, and the traditional recommendation system considers the recommendation problem as follows: given the historical behavior of a user, a list of items (item) which may be of interest to the user (the items include but are not limited to music, videos, articles, commodities and the like) is recommended to the user, the list is generated once, but the clicks of the items by the user are a time sequence, and the time sequence exists, so that the problems defined by the conventional recommendation system are not completely consistent with a real scene.

With the popularization of natural language processing, the serialization model is more mature, and since the language of the natural language processing field and the item of the recommendation system field are both a time sequence, the serialization model is more important in the recommendation system because it can naturally convert the recommendation problem into a sequence problem, namely: given the user's historical behavior, a sequence of items that the user is likely to interact in the future is predicted. Sequences are most different from lists in that sequences are chronological, while lists are ordered only by importance.

With reference to fig. 3, a schematic diagram of reply sentences are generated based on the Seq2Seq model of RNN, where the eq2Seq model is an architectural approach represented by an encoder (Encode) and a decoder (Decode), and the Seq2Seq model generates an output sequence Y from an input sequence X. In the seq2seq model represented by an encoder (Encode) which converts an input sequence into a vector of fixed length, and a decoder (Decode) which decodes the input vector of fixed length into an output sequence. It can be seen that, for the related art shown in fig. 3, the disadvantage of the project recommendation model based on the Seq2Seq model is that the model in the related art only establishes a one-to-one relationship for the training data Query-Response, and uses MLE to perform model optimization, which results in that the model generates many high-frequency general projects, so that the user cannot obtain more project recommendations according with the habits of the user in the vehicle-mounted environment. Therefore, the items in the generated sequence are high in popularity, high in occurrence frequency and too strong in universality, so that a high-quality text processing result is difficult to generate by the item recommendation model, the generation of item information is influenced, and the use experience of a user is also influenced.

To solve the defects in the related art, referring to fig. 12, fig. 12 is an optional flowchart of a project information processing method based on a project recommendation model according to an embodiment of the present invention, and fig. 13 is a structural schematic diagram of the project recommendation model according to the embodiment of the present invention, it can be understood that the steps shown in fig. 12 can be executed by various electronic devices operating a project recommendation model training apparatus, and the following description is made with respect to the steps shown in fig. 12.

Step 1201: acquiring project information in the vehicle-mounted using environment, and converting the project information into corresponding recognizable text information.

The historical behavior sequence of the user can be regarded as a short text, wherein each element in the sequence corresponds to a word in the text, and each element can respectively comprise the following components according to different use environments: user characteristics, commodity characteristics, and other characteristics such as cell phone model, gps location, etc.

Step 1202: determining at least one word-level hidden variable corresponding to the text information through an encoder of the item recommendation model.

Step 1203: generating, by a decoder of the project recommendation model, a text processing word corresponding to the word-level hidden variable and a selected probability of the text processing word according to the at least one word-level hidden variable.

In order to realize the diversity of the output results of the item recommendation model, the predicted item coding vector and the predicted user feature coding vector need to be obtained on the basis of the hidden variables of the preceding word level (i.e., the item coding vector of the previous round of prediction) at the decoding stage of the decoder of the item recommendation model.

In the traditional project recommendation model, the user characteristics are only considered during historical behavior encoding, and only item information is considered in the decoding stage. This will make the decoded item less and less affected by the user characteristics as the number of steps progresses. Therefore, additional user features are required to be added to increase different generation mechanisms of the decoding process for different users, and corresponding diversity is increased according to the characteristics of the users to enrich the generation results of the project recommendation model.

Further, in the decoding stage of the decoder network, an item (item) history sequence can be added into the decoder network, wherein the item history sequence is a list of items predicted by the item recommendation model, so that by means of an attention mechanism, when a new item is predicted, items not in the item list are strengthened, and the defects of items with high heat, high occurrence frequency and high universality in the generated sequence are avoided.

Further, in the decoding phase of the decoder network, an item diversity index may be added to the decoder network in the form of an objective function, where the item diversity requires that items have explicit attribute information to measure, such as different genres of music, different tags of video, different types of goods. If the information exists, the diversity index can be added into the objective function when the decoder network carries out decoding processing, so that the defects of items with high heat, high occurrence frequency and strong universality in the generated sequence are avoided.

Step 1204: selecting at least one text processing word to form a text processing result corresponding to the text information according to the selected probability of the text processing result;

step 1205: and converting the text processing result into new item information corresponding to the item recommendation model.

Taking a vehicle-mounted item recommendation model as an example, a use environment of the item information processing method of the item recommendation model provided by the application is described below, with reference to fig. 14 and 15, fig. 14 is a use scene schematic diagram of the item information processing method of the item recommendation model provided by the embodiment of the present invention, the item information processing method of the item recommendation model provided by the present invention may serve as a customer (packaged in a vehicle-mounted terminal or packaged in different mobile electronic devices) with a type that can be served in a form of a cloud service, and fig. 15 is a use scene schematic diagram of the item information processing method of the item recommendation model provided by the embodiment of the present invention, and the application is not particularly limited in a specific use scene, wherein the use scene is provided as a cloud service for enterprise customers to help the enterprise customers to train the item recommendation model according to different device use environments.

The beneficial technical effects are as follows:

compared with the Seq2Seq model used in the traditional technology, the generated sequence has the defects of high item popularity, high occurrence frequency and strong universality, so that the item recommendation model is difficult to generate high-quality text processing results, further the generation of item information is influenced, and the use experience of a user is also influenced.

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

Claims

1. A project information processing method based on a project recommendation model is characterized by comprising the following steps:

2. The method of claim 1, wherein the obtaining item information in an environment of use of the item recommendation model and converting the item information into corresponding recognizable text information comprises:

determining a dynamic noise threshold value matched with the use environment of the item recommendation model;

converting the project information into initial text information;

denoising the initial text information according to the dynamic noise threshold value, and triggering a dynamic word segmentation strategy matched with the dynamic noise threshold value;

and performing word segmentation processing on the initial text information according to a dynamic word segmentation strategy matched with the dynamic noise threshold value to form corresponding identifiable text information.

3. The method of claim 1, wherein the obtaining item information in an environment of use of the item recommendation model and converting the item information into corresponding recognizable text information comprises:

determining a fixed noise threshold that matches the usage environment of the item recommendation model;

converting the project information into initial text information;

denoising the initial text information according to the fixed noise threshold value, and triggering a fixed word segmentation strategy matched with the fixed noise threshold value;

and performing word segmentation processing on the initial text information according to a fixed word segmentation strategy matched with the fixed noise threshold value to form corresponding identifiable text information.

4. The method of claim 1, wherein generating, by the decoder network of the item recommendation model, from the at least one word-level hidden variable and the corresponding fused feature vector, a text processing word corresponding to the word-level hidden variable and a selected probability of the text processing word comprises:

determining a text processing result of the corresponding polling times according to the hidden variable of the at least one word level through a decoder network of the project recommendation model;

converting the text processing result of the corresponding polling times into a text processing result vector;

and generating text processing words corresponding to the hidden variables of the word level and the selected probability of the text processing words according to the text processing result vector and the fusion feature vector through a decoder network of the project recommendation model.

5. The method of claim 4, further comprising:

determining an item diversity function matching the item recommendation model;

and adjusting the output result of the decoder network of the project recommendation model through the project diversity function to realize that the text processing words and the selected probability of the text processing words are matched with the project diversity function.

6. The method of claim 1, further comprising:

acquiring user characteristics corresponding to the project recommendation model, and forming a user characteristic vector according to the user characteristics;

acquiring a text processing result of corresponding polling times, and converting the text processing result of the corresponding polling times into a feature vector of a predicted text processing result;

and fusing the user feature vector and the feature vector of the predicted text processing result to form a corresponding fused feature vector.

7. The method of claim 1, further comprising:

when the usage environment of the item recommendation model is a video recommendation process,

and adjusting parameters of a cyclic convolution neural network based on the multiple attention mechanism in the decoder network according to the fusion feature vector of the project recommendation model so as to realize that the parameters of the cyclic convolution neural network based on the multiple attention mechanism are matched with the fusion feature vector.

8. The method of claim 1, further comprising:

acquiring a training sample matched with the use environment of the project recommendation model;

extracting a feature set matched with the training sample through the item recommendation model;

and training the project recommendation model according to the feature set matched with the training sample and the corresponding target text label to determine model parameters matched with the project recommendation model.

9. The method according to any one of claims 1 to 8, further comprising:

and sending the item information, the parameter information of the item recommendation model and new item information which is generated by the item recommendation model and is matched with the use environment to a block chain network, so that nodes of the block chain network fill the item information, the parameter information of the item recommendation model and the new item information which is matched with the use environment into a new block, and when the new block is identified in common and consistent, the new block is added to the tail part of a block chain, so that the item recommendation model in different use environments of the same user can acquire the information in the block.

10. An item information processing apparatus based on an item recommendation model, the apparatus comprising:

11. The apparatus of claim 10,

12. The apparatus of claim 10,

13. The apparatus of claim 10,

14. An electronic device, characterized in that the electronic device comprises:

a memory for storing executable instructions;

a processor for implementing the method of processing item information based on the item recommendation model of any one of claims 1 to 9 when executing the executable instructions stored in the memory.

15. A computer-readable storage medium storing executable instructions, wherein the executable instructions, when executed by a processor, implement the item information processing method based on the item recommendation model according to any one of claims 1 to 9.