WO2022252636A1

WO2022252636A1 - Artificial intelligence-based answer generation method and apparatus, device, and storage medium

Info

Publication number: WO2022252636A1
Application number: PCT/CN2022/072187
Authority: WO
Inventors: 舒畅; 万欣茹; 张梓键; 陈又新
Original assignee: 平安科技（深圳）有限公司
Priority date: 2021-06-01
Filing date: 2022-01-14
Publication date: 2022-12-08
Also published as: CN113239169B; CN113239169A

Abstract

The present application relates to the field of artificial intelligence, and in particular, to an artificial intelligence-based answer generation method and apparatus, a device, and a storage medium, which enable a generation model to generate a more relevant and accurate answer when information of a sentence to be answered of a user is limited. The artificial intelligence-based answer generation method comprises: receiving a sentence to be answered, and inputting said sentence into a retrieval model; acquiring similar sentences screened out by the retrieval model from a preset dialogue corpus and matching said sentence, and acquiring answer sentences of the similar sentences in the dialogue corpus; on the basis of a preset semantic alignment keyword extraction model, extracting difference keywords between the similar sentences and said sentence to obtain a difference keyword set; extracting similar keywords of the answer sentences to obtain a similar keyword set; and inputting the difference keyword set and the similar keyword set into a trained generation model, and outputting an answer. By means of artificial intelligence-based natural language processing, a more relevant and accurate answer is given when the information of a question of a user is limited in an intelligent dialogue.

Description

Answer generating method, device, equipment and storage medium based on artificial intelligence

This application claims the priority of the Chinese patent application submitted to the China Patent Office on June 1, 2021, with the application number 202110609412.5, and the title of the invention is "Answer generation method, device, equipment and storage medium based on artificial intelligence", the entire content of which Incorporated in this application by reference.

technical field

The present application relates to the field of artificial intelligence, and in particular to an artificial intelligence-based answer generation method, device, equipment, and storage medium.

Background technique

With the continuous development of social media and the continuous advancement of artificial intelligence technology, intelligent dialogue systems play an increasingly important role in the current society. Question answering systems based on generative methods mostly use standard question-answer pairs as training data, and use generative models (usually the Encoder-Decoder framework) in natural language processing to train to obtain answer results. The method based on the generative model can automatically generate answers that are highly relevant to the user's questions. The inventors realized that due to the limited information contained in the user's questions, the results of the generative model may tend to be safe answers, such as "OK", " Hmm" and so on, this is meaningless and a waste of time for users of human-computer dialogue systems.

Contents of the invention

The present application provides an artificial intelligence-based answer generation method, device, device, and storage medium, which realizes giving more relevant and accurate answers in the case of limited user question information.

In a first aspect, the present application provides a method for generating an answer based on artificial intelligence, the method for generating an answer includes:

receiving the sentence to be replied, and inputting the sentence to be replied into the retrieval model;

Obtaining the similar sentences matched with the sentence to be replied selected by the retrieval model from the preset dialog corpus, and obtaining the answer sentence of the similar sentence in the dialog corpus;

Based on the preset semantic alignment keyword extraction model, extract the difference keywords between the similar sentence and the sentence to be replied, and obtain the difference keyword set;

Extracting similar keywords of the answer sentence to obtain a similar keyword set;

Input the difference keyword set and the similar keyword set into the trained generation model, and output the answer.

In a second aspect, the present application also provides an artificial intelligence-based answer generating device, the answer generating device comprising:

The sentence receiving module is used to receive the sentence to be replied, and input the sentence to be replied into the retrieval model;

A similar sentence acquisition module, used to obtain the retrieval model to filter similar sentences matching the sentences to be replied from the preset dialogue corpus, and obtain the answer sentences of the similar sentences in the dialogue corpus;

The difference keyword extraction module is used to extract the difference keywords between the similar sentence and the sentence to be replied based on the preset semantic alignment keyword extraction model, and obtain the difference keyword set;

A similar keyword extraction module is used to extract similar keywords of the answer sentence to obtain a similar keyword set;

An answer generating module, configured to input the difference keyword set and the similar keyword set into the trained generative question answering model, and output an answer.

In a third aspect, the present application also provides a computer device, the computer device includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and execute the The computer program realizes the answer generation method as described above.

In a fourth aspect, the present application further provides a storage medium, the storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the above answer generation method.

The answer generation method, device, computer equipment and storage medium disclosed in this application retrieve similar sentences from the user's sentences to be replied through the retrieval model, and obtain the answer sentences of similar sentences, further extract difference keywords for similar sentences, and extract Similar keywords, the difference keywords and similar keywords are used as the input of the generative model, which adds input information to the generative model. In this way, it is possible to overcome the generation model that gives safe and meaningless answers due to the limited information contained in the user's questions, and achieve more relevant and accurate answers in the case of limited information in the user's questions.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present application more clearly, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application. Ordinary technicians can also obtain other drawings based on these drawings on the premise of not paying creative work.

Fig. 1 is a schematic flowchart of an answer generation method provided by an embodiment of the present application;

Fig. 2 is a network schematic diagram of the answer generation method provided by the embodiment of the present application;

FIG. 3 is a schematic block diagram of an answer generation device provided by an embodiment of the present application;

FIG. 4 is a schematic structural block diagram of a computer device involved in an embodiment of the present application.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

The flow charts shown in the drawings are just illustrations, and do not necessarily include all contents and operations/steps, nor must they be performed in the order described. For example, some operations/steps can be decomposed, combined or partly combined, so the actual order of execution may be changed according to the actual situation.

It should be understood that the terms used in the specification of this application are for the purpose of describing specific embodiments only and are not intended to limit the application. As used in this specification and the appended claims, the singular forms "a", "an" and "the" are intended to include plural referents unless the context clearly dictates otherwise.

It should also be understood that the term "and/or" used in the description of the present application and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations .

With the continuous development of social media and the continuous advancement of artificial intelligence technology, intelligent dialogue systems play an increasingly important role in the current society. For example, Baidu’s Xiaodu artificial intelligence assistant and Microsoft’s dialogue system Xiaoice are all focused on realizing the business scenario of man-machine dialogue. The former uses a retrieval method to recall the user’s expected answer, while the latter uses a deep learning model. Automate answer generation based on user questions.

Question answering systems based on retrieval methods are mostly used in intelligent question answering systems in specific fields (such as Taobao platform robot customer service). Such methods usually maintain a dialogue data database that can be used for retrieval in advance, and use the user's input questions as questions. Then use the sentence similarity calculation to find the question sentence that is most similar to the user's question sentence, and then output the corresponding answer as a response. The question-answer pairs in the database have passed strict screening, so the answers returned to users are fluent in language and clear in expression. However, due to the limitations of the database scale and the sentence similarity matching method, the final answer is often irrelevant to the subject and the words are not expressive.

Question answering systems based on generative methods mostly use standard question and answer pairs as training data, and use generative models in natural language processing (usually the Encoder-Decoder framework) to train question answering systems. This method can automatically generate questions of the same height as user questions. However, due to the limited information contained in the user's questions, the results of the generative question answering model may tend to be safe answers, such as "okay", "um", etc. It's pointless and a waste of time.

Based on the defects and deficiencies of existing intelligent dialogue systems, this application proposes an answer generation method that combines the retrieval method and the generative method. The formula method can generate more relevant answers based on the user's sentence. Firstly, the user's sentence is retrieved based on the retrieval method, and the similar sentence matching the user's sentence and the corresponding answer are obtained. Then extract difference keywords for all similar sentences and user sentences, extract similar keywords for all replies, and use the obtained difference keywords and similar keywords as the input of the generative method, and the final answer is output by the generative model. In this way, it is possible to generate more accurate and relevant answers when the information provided by the user's statement is limited.

Embodiments of the present application provide an artificial intelligence-based answer generation method, device, device, and storage medium. Wherein, the method for generating an answer can be applied to a server, and can also be applied to a chip in a specific product, such as a chip in an artificial intelligence device. The retrieval model is used to add input information to the generation model based on the sentences to be replied, so that the answer given by the user is more accurate and relevant when the information given by the user is limited. Wherein, the server may be an independent server or a server cluster.

Some implementations of the present application will be described in detail below in conjunction with the accompanying drawings. In the case of no conflict, the following embodiments and features in the embodiments can be combined with each other.

Please refer to FIG. 1 . FIG. 1 is a schematic flowchart of an artificial intelligence-based answer generation method provided by an embodiment of the present application. The answer generation method can be applied to the server to obtain more relevant and accurate answers to the user's sentence.

As shown in FIG. 1 , the answer generation method specifically includes steps S101 to S105.

S101. Receive a sentence to be replied, and input the sentence to be replied into a retrieval model.

The sentences to be replied include question sentences or non-question sentences; they can be in the form of speech or text, and when the sentences to be replied are in speech form, they need to be converted into text form first to be input into the retrieval model.

S102. Obtain the similar sentences selected by the retrieval model from the preset dialog corpus that match the sentence to be replied, and obtain the answer sentences of the similar sentences in the dialog corpus.

The sentence to be replied and the sentence in the preset dialogue corpus are input into the retrieval model, and the Jaccard similarity coefficient for characterizing the similarity between the sentence in the preset dialogue corpus and the sentence to be replied is generated; Obtaining the similar sentences matched with the sentence to be replied based on the Jaccard similarity coefficient screened out from the preset dialogue corpus by the retrieval model; wherein, the similar sentence is similar to the Jaccard similarity coefficient of the sentence to be replied greater than the first preset threshold and less than the second preset threshold.

In some embodiments, after the sentences to be replied and the sentences in the preset dialogue prediction library are input into the retrieval model, the retrieval model generates a text similarity between the sentences in the preset dialogue corpus and the sentences to be replied to represent the sentences Jaccard similarity coefficient (Jaccard Similarity coefficient), and then judge whether the sentence is a similar sentence of the sentence to be replied according to the Jaccard similarity coefficient. The higher the Jaccard similarity coefficient value between sentences, the higher the similarity between sentences and the higher the matching degree. The retrieval model screens out the sentences whose Jaccard similarity coefficient value is greater than the first preset threshold and smaller than the second preset threshold in the preset dialogue corpus and the sentence to be replied, and selects the Jaccard similarity from the filtered results according to the requirements. The preset number with the highest coefficient value is used as the final output similar sentences, and then the corresponding answer sentences of these similar sentences in the preset dialogue prediction library are also obtained, and the similar sentences output by the retrieval model and the corresponding answer sentences are one by one Correspond to form a similar question and answer pair.

Exemplarily, the retrieval model screens out sentences in the preset dialogue corpus that have a Jaccard similarity coefficient greater than 0.5 and smaller than 0.9 with the sentence to be replied, and selects 3 sentences with the highest Jaccard similarity coefficients as similar sentences to output.

It should be noted that the first preset threshold, the second preset threshold, and the preset number can be set according to actual applications, which is not specifically limited in this application.

In some other embodiments, the retrieval model can also be used to evaluate the text similarity between sentences based on the cosine similarity, so as to match similar sentences similar to the sentence to be replied. The cosine similarity is to evaluate the similarity of two sentence vectors by calculating the angle between the two sentence vectors.

It should be noted that the preset dialogue corpus is a pre-maintained database containing dialogue sentences, including both question and answer sentences and answer sentences corresponding to the question and answer sentences. The dialogue corpus can be based on the requirements of the application scenario, and can be a dialogue corpus that collects general daily conversations, or a dialogue corpus that collects specific application scenarios.

S103. Based on the preset semantic alignment keyword extraction model, extract the difference keywords between the similar sentence and the sentence to be replied, and obtain a difference keyword set.

This application proposes a Semantic Alignment Keyword Extraction (SAKE) model for keyword extraction. The semantic alignment keyword extraction model is based on the use of cross-attention (Cross-Attention) to calculate the attention matrix representing deep semantics, and extract keywords based on the attention matrix.

Specifically, the sentence to be replied is converted into the first vector, the similar sentence is converted into the second vector, and the attention matrix is calculated according to the first vector and the second vector; the minimum value of the attention matrix is obtained to obtain the similar sentence and the waiting vector The difference keywords of the reply sentence; the difference keywords are combined to obtain the difference keyword set.

First of all, the text sentence needs to be converted into a vector form, and there are many methods for sentence vectorization.

In some embodiments, the text sentence is firstly one-hot encoded to become a sparse matrix, and then the embedding layer is randomly initialized, the sparse matrix is transformed into a dense matrix through linear transformation, and the mutually independent one-hot vectors become A relational vector that considers the internal connection between words is proposed.

In other embodiments, the sentence vector can also be generated based on a neural network language model (Nerual Network Language Model, NNLM).

It should be noted that all similar sentences need to be aligned to the sentences to be replied, so the sentences to be replied are converted into the first vector, and other similar sentences are converted into the second vector, and the attention between each similar sentence and the sentence to be replied is calculated in turn. force matrix. Among them, the attention matrix is expressed as:

Wherein, ν _A is the sentence to be replied converted into the first vector, ν _B is the similar sentence converted into the second vector, and d _k is the normalization coefficient.

By calculating the minimum value of the attention matrix, the words that exist in the sentence to be replied but not in the similar sentence can be obtained, which are the difference keywords between the two. Wherein, the difference keyword formula is as follows:

Among them, i represents the row, j represents the column, L is a predetermined hyperparameter,

Indicates the row and the smallest L rows, corresponding to the extracted difference keywords.

Take the union of the difference keywords of all similar sentences and the sentences to be replied, and obtain the difference keyword set.

Exemplarily, for example, the sentence to be replied is: Which is better, wine or white wine? Assume that the similar sentences obtained through the retrieval model are: Similar sentence 1: Is liquor better than wine? Similar sentence 2: Is there any wine that tastes better than liquor? The difference keywords between similar statement 1 and the statement to be replied are: "more", "and" and "which". The difference keywords between the similar statement 2 and the statement to be replied are: "wine", "and", "which". Take the union of the difference keywords to get the difference keyword set: ["more", "and", "which", "wine"].

S104. Extract similar keywords of the answer sentence to obtain a similar keyword set.

Specifically, the first answer sentence in the answer sentences is converted into a first answer vector, the other answer sentences in the answer sentences are converted into a second answer vector, and an attention matrix is calculated according to the first answer vector and the second answer vector. Wherein, the first answer sentence is the answer sentence corresponding to the similar sentence with the highest matching degree with the sentence to be replied.

It should be noted that: the answer sentence corresponding to the similar sentence that best matches the sentence to be replied is used as a reference answer sentence, and other answer sentences need to be aligned with the reference answer sentence. Therefore, the reference answer sentence is converted into the first answer vector, other answer sentences in the answer sentence are converted into the second answer vector, and the attention matrix between other answer sentences and the reference answer sentence is calculated in sequence. Calculate the maximum value of the attention matrix to obtain similar keywords between the first answer sentence and other answer sentences in the answer sentence. Wherein, the similar keyword formula is as follows:

Wherein, the similar keyword formula is as follows:

Among them, i represents the row, j represents the column, and L is a predetermined hyperparameter, which represents the row and the largest L rows, corresponding to the extracted similar keywords.

All other answer sentences other than the reference answer sentence and the reference answer sentence are used to obtain similar keywords, and all similar keywords are combined to obtain a similar keyword set.

Exemplarily, for example, the answer statement 1 is "I personally don't like drinking alcohol", and the answer statement 2 is "depending on personal taste, I prefer white wine". The similar keywords of answer statement 2 and answer statement 1 are: "personal", "like" and "wine". Therefore, the similar keyword set is: ["individual", "like" and "wine"].

The keyword extraction model based on semantic alignment proposed in this application uses the attention matrix for normalized accumulation, realizing the consideration of deep semantic correlation and cumulative semantic characteristics in keyword extraction, so that the extracted keywords are more accurate.

S105. Input the difference keyword set and the similar keyword set into the trained generation model, and output an answer. The generation model includes a trained first generation model and a trained second generation model.

Specifically, input the difference keyword set and the similar keyword set into the trained first generation model, and output an ordered keyword sequence by predicting the position of the words in the difference keyword set and the similar keyword set in the sentence; The ordered keyword sequence is input into the trained second generation model, and the answer is output.

In the embodiment provided in this application, both the first generation model and the second generation model are based on the pre-trained language model BERT (Bidirectional Encoder Representation from Transformers), but they are trained based on different model parameters and training strategies. The BERT model no longer uses the traditional one-way language model or the method of shallow splicing two one-way language models for pre-training, but uses a new masked language model (masked language model, MLM) to generate a deep two-way language characterization. The BERT model uses MLM to pre-train bidirectional Transformers to generate deep bidirectional language representations. After pre-training, only need to add an additional output layer for fine-tuning (fine-tune), can achieve high-level performance in a variety of downstream tasks, and does not require task-specific structural modifications to the BERT model.

In the embodiment provided in this application, the network structure of the first generation model is: the first BERT model followed by the first fully connected layer and Softmax. The training objective of the first generation model is to output a predicted ordered word set based on a given unordered word set. Specifically, input the unordered keyword set into the first BERT model, output the enhanced semantic vector of the word through the transformer encoder, connect the first fully connected layer and Softmax layer after the first BERT model, and output the predicted ordered keywords set. It should be noted that during the training process of the first generative model, the parameters of the first BERT model are not fine-tuned, but the parameters of the first fully connected layer are adjusted to obtain the trained first generative model.

In the embodiment provided in this application, the network structure of the second generation model is: the second BERT model followed by the second fully connected layer and Softmax. The training objective of the second generation model is to output a predicted complete sentence based on the ordered keyword set. The second generative model is based on contextual information and semantic understanding. In the output sentence, words not included in the keyword set can be obtained through prediction, and some words included in the keyword set can also be ignored. It should be noted that during the training process of the second generation model, the parameters of the second BERT model and the second fully connected layer need to be fine-tuned to obtain a trained second generation model.

First, the unordered keyword set obtained by taking the union of the difference keyword set and the similar keyword set is input into the trained first generative model to generate an ordered keyword set, where the ordered keyword set is equivalent to Combine discrete words into an incomplete sentence. Secondly, the ordered keyword set is input into the trained second generation model, and the second generation model outputs a complete sentence based on the ordered keyword set.

The predicted answer is output through a two-stage generative model, in which the first generative model predicts the location information to obtain an ordered discrete word set, and the second stage predicts the final generated answer sentence based on the ordered discrete word set, making the generated result more fluent.

In the answer generation method provided by this application, firstly, based on the retrieval model, similar sentences are retrieved for the user's waiting sentences, and the answer sentences of the similar sentences are obtained, and the difference keywords are further extracted for the similar sentences, similar keywords are extracted for the answer sentences, and the difference keywords are extracted. Words and similar keywords are used as the input of the generative model, which adds input information to the generative model. In this way, it is possible to overcome the problem that the generative model gives safe and meaningless answers due to the limited information contained in the user's questions, and realize the result of giving more relevant and accurate answers in the case of limited information in the user's questions.

Referring to FIG. 3 , an embodiment of the present application provides a schematic block diagram of an answer generation device, and the answer generation device may be configured in a server.

As shown in FIG. 3 , the answer generation device 400 includes: a sentence receiving module 401 , a similar sentence obtaining module 402 , a difference keyword extraction module 403 , a similar keyword extraction module 404 , and an answer generation module 405 .

The sentence receiving module 401 is used to receive the sentence to be replied, and input the sentence to be replied into the retrieval model;

The similar sentence obtaining module 402 is configured to obtain the similar sentence matched with the sentence to be replied selected from the preset dialogue corpus by the retrieval model, and obtain the answer sentence of the similar sentence in the dialogue corpus.

The difference keyword extraction module 403 is configured to extract the difference keywords between the similar sentence and the sentence to be replied based on a preset semantic alignment keyword extraction model to obtain a difference keyword set.

The similar keyword extraction module 404 is configured to extract similar keywords of the answer sentence to obtain a similar keyword set.

The answer generation module 405 is configured to input the difference keyword set and the similar keyword set into the trained generation model, and output an answer.

It should be noted that those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described device and each module can refer to the corresponding process in the foregoing method embodiment, which is not repeated here. repeat.

The above-mentioned apparatus can be realized in the form of a computer program, and the computer program can be run on the computer device as shown in FIG. 4 .

Referring to FIG. 4 , FIG. 4 is a schematic structural block diagram of a computer device provided by an embodiment of the present application. The computer device may be a server.

Referring to FIG. 4 , the computer device includes a processor, a memory, and a network interface connected through a system bus, wherein the memory may include a storage medium and an internal memory.

The storage medium may store operating systems and computer programs. The computer program includes program instructions which, when executed, cause a processor to perform any one of the answer generation methods.

The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

The internal memory provides an environment for running the computer program in the storage medium. When the computer program is executed by the processor, the processor can execute any answer generating method.

This network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art can understand that the structure shown in Figure 4 is only a block diagram of a part of the structure related to the solution of this application, and does not constitute a limitation on the computer equipment to which the solution of this application is applied. The specific computer equipment can be More or fewer components than shown in the figures may be included, or some components may be combined, or have a different arrangement of components.

It should be understood that the processor may be a central processing unit (Central Processing Unit, CPU), and the processor may also be other general processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), Field-Programmable Gate Array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Wherein, the general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

Wherein, in one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

Extracting difference keywords from the similar sentence and the sentence to be replied based on a preset semantic alignment keyword extraction model to obtain a difference keyword set;

Extracting similar keywords to the answer sentence to obtain a similar keyword set;

In one embodiment, when the processor acquires similar sentences that match the sentences to be replied and is screened out by the retrieval model from a preset dialogue corpus, it is used to realize:

The sentence to be replied and the sentence in the preset dialogue corpus are input into the retrieval model, and the Jaccard similarity coefficient for characterizing the similarity between the sentence in the preset dialogue corpus and the sentence to be replied is generated;

Obtaining the similar sentences matched with the sentence to be replied based on the Jaccard similarity coefficient screened out from the preset dialogue corpus by the retrieval model; wherein, the similar sentence is similar to the Jaccard similarity coefficient of the sentence to be replied greater than the first preset threshold and less than the second preset threshold.

In one embodiment, when the processor realizes the keyword extraction model based on the preset semantic alignment, extracts the difference keywords between the similar sentence and the sentence to be replied, and obtains the difference keyword set, it is used to realize:

Converting the sentence to be replied into a first vector, converting a similar sentence into a second vector, and calculating an attention matrix according to the first vector and the second vector;

Find the minimum value to described attention matrix, obtain the difference keyword of similar sentence and sentence to be replied;

The difference keywords are combined to obtain a difference keyword set.

In one embodiment, when the processor is used to calculate the attention matrix, it is used to realize:

Wherein, ν _A is that the sentence to be replied is converted into the first vector, and ν _B is converted into the second vector for the similar sentence, and d _k is a normalization coefficient;

The difference keyword formula is as follows:

Among them, i represents the row, j represents the column, and L is a predetermined hyperparameter, which means seeking

row and the smallest L rows, corresponding to the extracted difference keywords.

In one embodiment, when the processor is used to extract similar keywords of the answer sentence to obtain a similar keyword set, it is used to realize:

Convert the first answer sentence of the answer sentence into a first answer vector, convert other answer sentences in the answer sentence into a second answer vector, and calculate attention according to the first answer vector and the second answer vector A matrix; wherein, the first answer sentence is the answer sentence corresponding to the similar sentence with the highest matching degree with the sentence to be replied;

Finding the maximum value of the attention matrix to obtain similar keywords between the first answer sentence and other answer sentences in the answer sentence;

Take the union of all similar keywords to obtain a similar keyword set;

Wherein, the similar keyword formula is as follows:

Among them, i represents the row, j represents the column, and L is a predetermined hyperparameter, which means seeking the row

and the largest L rows, corresponding to the extracted similar keywords.

In one embodiment, the trained generative model includes a trained first generative model and a trained second generative model, and the processor is used to implement the difference keyword set and the similar key The word set is input to the trained generation model, and when outputting the answer, it is used to realize:

Input the difference keyword set and the similar keyword set into the trained first generation model, and output ordered keyword sequence;

Input the ordered keyword sequence into the trained second generation model, and output an answer.

The trained generation model is obtained based on the pre-trained language model BERT.

Embodiments of the present application also provide a storage medium, the storage medium stores a computer program, the computer program includes program instructions, and the processor executes the program instructions to implement any Item answer generation method.

Wherein, the storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as a hard disk or a memory of the computer device. The storage medium may also be an external storage device of the computer device, such as a plug-in hard disk equipped on the computer device, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, Flash card (Flash Card), etc. The storage medium may be non-volatile or volatile.

The above is only a specific embodiment of the application, but the scope of protection of the application is not limited thereto. Any person familiar with the technical field can easily think of various equivalents within the scope of the technology disclosed in the application. Modifications or replacements, these modifications or replacements shall be covered within the scope of protection of this application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

A method for generating answers based on artificial intelligence, including:

receiving the sentence to be replied, and inputting the sentence to be replied into the retrieval model;

Obtaining the similar sentences matched with the sentence to be replied selected by the retrieval model from the preset dialog corpus, and obtaining the answer sentence of the similar sentence in the dialog corpus;

Based on the preset semantic alignment keyword extraction model, extract the difference keywords between the similar sentence and the sentence to be replied, and obtain the difference keyword set;

Extracting similar keywords of the answer sentence to obtain a similar keyword set;

Input the difference keyword set and the similar keyword set into the trained generation model, and output the answer.
The answer generating method according to claim 1, wherein said acquiring similar sentences matched with said sentences to be replied from a preset dialogue corpus screened by said retrieval model comprises:

The sentence to be replied and the sentence in the preset dialogue corpus are input into the retrieval model, and the Jaccard similarity coefficient for characterizing the similarity between the sentence in the preset dialogue corpus and the sentence to be replied is generated;

Obtaining the similar sentences matched with the sentence to be replied selected from the preset dialogue corpus based on the Jaccard similarity coefficient by the retrieval model;

Wherein, the Jaccard similarity coefficient between the similar sentence and the sentence to be replied is greater than a first preset threshold and smaller than a second preset threshold.
The answer generation method according to claim 1, wherein the keyword extraction model based on the preset semantic alignment extracts the difference keywords between the similar sentence and the sentence to be replied, and obtains the difference keyword set comprising:

Converting the sentence to be replied into a first vector, converting a similar sentence into a second vector, and calculating an attention matrix according to the first vector and the second vector;

Find the minimum value to described attention matrix, obtain the difference keyword of similar sentence and sentence to be replied;

The difference keywords are combined to obtain a difference keyword set.
The answer generation method according to claim 3, wherein the attention matrix is expressed as:

Wherein, ν A is the first vector converted from the sentence to be replied, and ν B is the second vector converted from the similar sentence, and d k is a normalization coefficient;

The difference keyword formula is as follows:

Among them, i represents the row, j represents the column, L is a predetermined hyperparameter,
Indicates the row and the smallest L rows, corresponding to the extracted difference keywords.
The answer generating method according to claim 4, wherein said extracting similar keywords of said answer sentence, and obtaining a similar keyword set includes:

Convert the first answer sentence of the answer sentence into a first answer vector, convert other answer sentences in the answer sentence into a second answer vector, and calculate attention according to the first answer vector and the second answer vector A matrix; wherein, the first answer sentence is the answer sentence corresponding to the similar sentence with the highest matching degree with the sentence to be replied;

Finding the maximum value of the attention matrix to obtain similar keywords between the first answer sentence and other answer sentences in the answer sentence;

Take the union of all similar keywords to obtain a similar keyword set;

Wherein, the similar keyword formula is as follows:

Among them, i represents the row, j represents the column, L is a predetermined hyperparameter,
Indicates the row and the largest L rows, corresponding to the extracted similar keywords.
The answer generation method according to claim 1, wherein the trained generation model includes a trained first generation model and a trained second generation model;

The step of inputting the difference keyword set and the similar keyword set into the trained generation model, and outputting an answer includes:

Input the difference keyword set and the similar keyword set into the trained first generation model, and output ordered keyword sequence;

Input the ordered keyword sequence into the trained second generation model, and output an answer.
The answer generation method according to claim 6, wherein the trained generation model is obtained based on a pre-trained language model BERT.
A device for generating answers based on artificial intelligence, wherein the device includes:

The sentence receiving module is used to receive the sentence to be replied, and input the sentence to be replied into the retrieval model;

A similar statement acquiring module, configured to acquire the similar statements matched with the statement to be replied selected by the retrieval model from the preset dialogue corpus, and acquire the answer statement of the similar statement in the dialogue corpus;

The difference keyword extraction module is used to extract the difference keywords between the similar sentence and the sentence to be replied based on the preset semantic alignment keyword extraction model, and obtain the difference keyword set;

A similar keyword extraction module is used to extract similar keywords of the answer sentence to obtain a similar keyword set;

An answer generation module, configured to input the difference keyword set and the similar keyword set into a trained generation model, and output an answer.
A computer device, wherein the computer device includes a memory and a processor;

The memory is used to store computer programs;

The processor is configured to execute the computer program and implement the following steps when executing the computer program: receiving a sentence to be replied, and inputting the sentence to be replied into a retrieval model;

Obtaining the similar sentences matched with the sentence to be replied selected by the retrieval model from the preset dialog corpus, and obtaining the answer sentence of the similar sentence in the dialog corpus;

Based on the preset semantic alignment keyword extraction model, extract the difference keywords between the similar sentence and the sentence to be replied, and obtain the difference keyword set;

Extracting similar keywords of the answer sentence to obtain a similar keyword set;

Input the difference keyword set and the similar keyword set into the trained generation model, and output the answer.
The computer device according to claim 9, wherein the processor realizes the step of obtaining the similar sentences matched with the sentence to be replied selected from the preset dialogue corpus by the retrieval model, including:

The sentence to be replied and the sentence in the preset dialogue corpus are input into the retrieval model, and the Jaccard similarity coefficient for characterizing the similarity between the sentence in the preset dialogue corpus and the sentence to be replied is generated;

Obtaining the similar sentences matched with the sentence to be replied selected from the preset dialogue corpus based on the Jaccard similarity coefficient by the retrieval model;

Wherein, the Jaccard similarity coefficient between the similar sentence and the sentence to be replied is greater than a first preset threshold and smaller than a second preset threshold.
The computer device according to claim 9, wherein the processor implements a preset semantic alignment keyword extraction model to extract the difference keywords between the similar sentence and the sentence to be replied, and obtain the difference keyword set steps, including:

Converting the sentence to be replied into a first vector, converting a similar sentence into a second vector, and calculating an attention matrix according to the first vector and the second vector;

Find the minimum value to described attention matrix, obtain the difference keyword of similar sentence and sentence to be replied;

The difference keywords are combined to obtain a difference keyword set.
The computer device according to claim 9, wherein said processor implements the step of calculating the attention matrix, comprising:

The attention matrix is expressed as:

Wherein, ν A is the first vector converted from the sentence to be replied, and ν B is the second vector converted from the similar sentence, and d k is a normalization coefficient;

The difference keyword formula is as follows:

Among them, i represents the row, j represents the column, L is a predetermined hyperparameter,
Indicates the row and the smallest L rows, corresponding to the extracted difference keywords.
The computer device according to claim 9, wherein said processor implements the step of extracting similar keywords of said answer sentence to obtain a similar keyword set, including:

Convert the first answer sentence of the answer sentence into a first answer vector, convert other answer sentences in the answer sentence into a second answer vector, and calculate attention according to the first answer vector and the second answer vector A matrix; wherein, the first answer sentence is the answer sentence corresponding to the similar sentence with the highest matching degree with the sentence to be replied;

Finding the maximum value of the attention matrix to obtain similar keywords between the first answer sentence and other answer sentences in the answer sentence;

Take the union of all similar keywords to obtain a similar keyword set;

Wherein, the similar keyword formula is as follows:

Among them, i represents the row, j represents the column, L is a predetermined hyperparameter,
Indicates the row and the largest L rows, corresponding to the extracted similar keywords.
The computer device according to claim 9, wherein the trained generation model includes a trained first generation model and a trained second generation model, and the processor realizes combining the difference keyword set and the The steps of inputting the trained generative model of the similar keyword set and outputting the answer include:

Input the difference keyword set and the similar keyword set into the trained first generation model, and output ordered keyword sequence;

Input the ordered keyword sequence into the trained second generation model, and output an answer;

Wherein, the trained generation model is obtained based on the pre-trained language model BERT.
A storage medium, wherein the storage medium stores a computer program, the storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the following steps:

receiving the sentence to be replied, and inputting the sentence to be replied into the retrieval model;

Obtaining the similar sentences matched with the sentence to be replied selected by the retrieval model from the preset dialog corpus, and obtaining the answer sentence of the similar sentence in the dialog corpus;

Based on the preset semantic alignment keyword extraction model, extract the difference keywords between the similar sentence and the sentence to be replied, and obtain the difference keyword set;

Extracting similar keywords of the answer sentence to obtain a similar keyword set;

Input the difference keyword set and the similar keyword set into the trained generation model, and output the answer.
The storage medium according to claim 15, wherein the processor realizes the step of obtaining the similar sentences matched with the sentence to be replied selected from the preset dialogue corpus by the retrieval model, including:

The sentence to be replied and the sentence in the preset dialogue corpus are input into the retrieval model, and the Jaccard similarity coefficient for characterizing the similarity between the sentence in the preset dialogue corpus and the sentence to be replied is generated;

Obtaining the similar sentences matched with the sentence to be replied selected from the preset dialogue corpus based on the Jaccard similarity coefficient by the retrieval model;

Wherein, the Jaccard similarity coefficient between the similar sentence and the sentence to be replied is greater than a first preset threshold and smaller than a second preset threshold.
The computer device according to claim 15, wherein the processor implements a preset semantic alignment keyword extraction model to extract the difference keywords between the similar sentence and the sentence to be replied, and obtain the difference keyword set steps, including:

Converting the sentence to be replied into a first vector, converting a similar sentence into a second vector, and calculating an attention matrix according to the first vector and the second vector;

Find the minimum value to described attention matrix, obtain the difference keyword of similar sentence and sentence to be replied;

The difference keywords are combined to obtain a difference keyword set.
The computer device according to claim 15, wherein said processor implements the step of calculating an attention matrix, comprising:

The attention matrix is expressed as:

Wherein, ν A is the first vector converted from the sentence to be replied, and ν B is the second vector converted from the similar sentence, and d k is a normalization coefficient;

The difference keyword formula is as follows:

Among them, i represents the row, j represents the column, L is a predetermined hyperparameter,
Indicates the row and the smallest L rows, corresponding to the extracted difference keywords.
The computer device according to claim 15, wherein said processor implements the step of extracting similar keywords of said answer sentence to obtain a similar keyword set, including:

Convert the first answer sentence of the answer sentence into a first answer vector, convert other answer sentences in the answer sentence into a second answer vector, and calculate attention according to the first answer vector and the second answer vector A matrix; wherein, the first answer sentence is the answer sentence corresponding to the similar sentence with the highest matching degree with the sentence to be replied;

Finding the maximum value of the attention matrix to obtain similar keywords between the first answer sentence and other answer sentences in the answer sentence;

Take the union of all similar keywords to obtain a similar keyword set;

Wherein, the similar keyword formula is as follows:

Among them, i represents the row, j represents the column, L is a predetermined hyperparameter,
Indicates the row and the largest L rows, corresponding to the extracted similar keywords.
The computer device according to claim 15, wherein the trained generation model includes a trained first generation model and a trained second generation model, and the processor realizes combining the difference keyword set and the The steps of inputting the trained generative model of the similar keyword set and outputting the answer include:

Input the difference keyword set and the similar keyword set into the trained first generation model, and output ordered keyword sequence;

Input the ordered keyword sequence into the trained second generation model, and output an answer;

Wherein, the trained generation model is obtained based on the pre-trained language model BERT.