CN117474505A - Conversational part-part recruitment method based on local knowledge base and AI large model - Google Patents
Conversational part-part recruitment method based on local knowledge base and AI large model Download PDFInfo
- Publication number
- CN117474505A CN117474505A CN202311178181.2A CN202311178181A CN117474505A CN 117474505 A CN117474505 A CN 117474505A CN 202311178181 A CN202311178181 A CN 202311178181A CN 117474505 A CN117474505 A CN 117474505A
- Authority
- CN
- China
- Prior art keywords
- information
- user
- knowledge base
- local knowledge
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007115 recruitment Effects 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000001514 detection method Methods 0.000 claims description 9
- 238000012790 confirmation Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 3
- 238000004891 communication Methods 0.000 abstract description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 241000157593 Milvus Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- INFDPOAKFNIJBF-UHFFFAOYSA-N paraquat Chemical compound C1=C[N+](C)=CC=C1C1=CC=[N+](C)C=C1 INFDPOAKFNIJBF-UHFFFAOYSA-N 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/105—Human resources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Human Resources & Organizations (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Human Computer Interaction (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a dialogue type part-time recruitment method based on a local knowledge base and an AI large model, which comprises the following steps: based on information input by a user, accurately capturing the intention of the user by using an AI large model; customizing a workflow for each intention according to the disagreement graph of the user, and calling a local knowledge base in the workflow to assist in completing the recruitment of the part-time job. The invention can provide good question-answer feedback for the job seeker, is convenient for the job seeker to know the information of the part-time job, and can improve the recruitment efficiency for recruitment and reduce invalid communication.
Description
Technical Field
The invention relates to the field of part job recruitment application, in particular to a conversational part job recruitment method based on a local knowledge base and an AI large model.
Background
The recruitment industry is currently undergoing rapid development and transformation, and on one hand, with the surge of the Internet and the civilian entrepreneur, all industries face industry upgrades. The current AI large model is an important product for the development of modern artificial intelligence. AI large models are roughly divided into two main categories, one category is to conduct fine adjustment according to industry data imported by an open source model, for example, a new model is exported based on fine adjustment between meaning thousands; the other is based on large models of domestic commercialization, such as Confucius, xinfei star fire and ChatGLM. The existing AI large model is basically a chat large model, and is characterized in that the large model can answer according to the requirements in the prompt words after a user inputs a problem (namely the prompt words). However, the answer of the AI large model is limited to the data trained by the AI large model, and is lack of other specific answers, so that the AI large model is difficult to be combined with the actual industry. Meanwhile, the maximum point of the part-time recruitment business is different from the full-time recruitment business in that the full-time recruitment has special personnel to process recruitment matters, and the part-time recruitment business has no special personnel to follow up the recruitment due to the characteristics of short-term and temporary requirements and the like. This also results in often no way to directly chat with the employer when the user wants to ask for some information about the job. Therefore, how to use the large AI model to realize the recruitment of the part-time job network and improve the recruitment efficiency become the technical problem to be solved urgently by the applicant.
Disclosure of Invention
The invention aims to provide a conversational part-time recruitment method based on a local knowledge base and an AI large model. The invention can provide good question-answer feedback for the job seeker, is convenient for the job seeker to know the information of the part-time job, and can improve the recruitment efficiency for recruitment and reduce invalid communication.
In order to solve the technical problems, the technical scheme provided by the invention is as follows: a dialogue type part job recruitment method based on a local knowledge base and an AI large model comprises the following steps: based on information input by a user, accurately capturing the intention of the user by using an AI large model;
customizing a workflow for each intention according to the disagreement graph of the user, and calling a local knowledge base in the workflow to assist in completing the recruitment of the part-time job.
The local knowledge base and the dialogue type part job recruitment method of the AI large model, wherein the local database comprises a knowledge graph and a vector database;
the knowledge graph is formed by carrying out entity identification on part-time information, and carrying out information classification and relation association by utilizing industry classification and Embedding similarity to form the knowledge graph;
the vector database is formed by selecting a pyrtorch version of an open-source bert-base-Chinese model to perform an encoding process on the part-time text information and then storing the part-time text information in the database.
The aforementioned conversational part-time recruitment method of local knowledge base and AI large model, the entity identification shows that selecting user search hot word, city, business district, subway, school and post category data to make detailed labeling and then using as industry word base, then using hanNLP tool to make named entity identification, and storing the associated knowledge graph data into open-source graph database neo4j
The aforementioned conversational part-time recruitment method for the local knowledge base and the AI large model is characterized in that: the method for accurately capturing the intention of the user by using the AI large model based on the information input by the user comprises the following steps:
the method comprises the steps of performing pre-layer processing on characters or voices input by a user, converting the voices into characters, detecting and intercepting sensitive words, correcting wrongly written characters and/or removing special characters in translation mode, then assembling the processed input information into prompt words, and then transmitting the assembled prompt words into an AI large model for intention recognition and obtaining results.
The aforementioned conversational part job recruitment methods of the local knowledge base and AI big model, the intent recognition includes find work intent, query post information intent, query interview arrangement intent, and other intents.
The aforementioned conversational part-time recruitment method of the local knowledge base and the AI big model, aiming at finding the work intention, the workflow is to check the user resume information first, if it is complete, the local knowledge is utilized to identify the input information; if the information is incomplete, guiding the user to perfect the necessary filling information through multiple rounds of dialogue, then utilizing the local knowledge to carry out entity identification on the input information, then carrying out search recommendation from a local knowledge base, and finally outputting recommendation post information according to a template format.
The method for achieving the dialog part job recruitment of the local knowledge base and the AI large model aims at inquiring the position information intention, the task bit stream is used for firstly combining dialog context to take out user input, then combining recommended position information to complete prompt word assembly, then transmitting the assembled prompt word into the AI large model, calling the local knowledge base to answer, and meanwhile carrying out security detection on answer content.
According to the dialogue type part job recruitment method of the local knowledge base and the AI large model, aiming at inquiring interview arrangement intention, the industrial bit stream is used for completing prompt word assembly according to interview time filled in when a merchant issues posts in the local knowledge base, the assembled prompt word is transmitted into the AI large model, the AI large model answers a user, meanwhile, safety detection is carried out on answer content, after the job seeker confirms, the answer content is sent to the merchant to confirm in a short message or other channel mode, and after secondary confirmation of the merchant, a prompt is returned, otherwise, the job seeker or the merchant waits for confirmation again.
According to the conversational part-time recruitment method of the local knowledge base and the AI large model, the workflow is used for updating the local knowledge base according to feedback information and optimizing prompting words and flow of the AI large model aiming at other intents.
Compared with the prior art, the invention builds a brand-new intelligent part-time recruitment service flow, can accurately capture the intention of the user from the voice and text information input by the user by means of the capability of the AI large model, and can customize the workflow aiming at each intention according to the disagreement diagram of the user. In finding the working intention, the invention can guide the user to perfect the needed resume information by using a preset multi-round dialogue, and extract and store the information by using a local part-time information knowledge base. Meanwhile, the invention can rapidly and accurately recommend the part-time information meeting the requirements to the user by utilizing the local part-time information knowledge base, and can rapidly answer the questions about the post details of the user according to the post information set by the employer. In addition, the invention can intelligently coordinate interview and job scheduling of the user and the employer, not only improves the efficiency, but also can collect service evaluation of the user and the employer to form closed loop feedback. On one hand, the invention brings brand new part-time experience to the user through a conversational job seeking mode, and the intention recognition capability of the AI large model and the local knowledge base can quickly help the user to find an ideal post; on the other hand, the invention integrates the information of the user and the employer by means of the AI large model, intelligently arranges interviews and job entries, saves a great amount of resources and cost, and improves the service efficiency of the part-time job. After the end of the part-time, the user and employer can make service feedback that will be applied in the subsequent improvement of the intelligent process to form a service closed loop.
Drawings
FIG. 1 is a schematic diagram of the local knowledge base of the present invention
Fig. 2 is a schematic flow chart of the present invention.
Detailed Description
The invention is further illustrated by the following examples and figures, which are not intended to be limiting.
Examples: a dialogue type part job recruitment method based on a local knowledge base and an AI large model comprises the following steps:
based on information input by a user, accurately capturing the intention of the user by using an AI large model;
customizing a workflow for each intention according to the disagreement graph of the user, and calling a local knowledge base in the workflow to assist in completing the recruitment of the part-time job.
In this embodiment, as shown in fig. 1, the local database includes a knowledge graph and a vector database;
the knowledge graph is formed by carrying out entity identification on part-time information, and carrying out information classification and relation association by utilizing industry classification and Embedding similarity to form the knowledge graph; the entity identification shows that the user is selected to search hot words, cities, business circles, subways, schools and post category data, the data are marked in detail and then used as industry word stock, then a hanNLP tool is used for named entity identification, and the associated knowledge graph data are stored in an open-source graph database neo4 j.
The vector database is formed by selecting a pytorch version of an open-source bert-base-Chinese model to perform an Embedding process on part-time text information (such as a position title, a position description, a working place and the like) and then storing the part-time text information in a database (such as Milvus) to form the vector database.
The AI big model selected in this embodiment is a chat big model (Xingfeixing fire) and is characterized in that the big model answers according to the requirements in the prompt words after the user inputs the questions (i.e. the prompt words). In this embodiment, the intent recognition is performed by using the large model according to the chat context of the user and the chat color is performed based on post information, and when in use, the information in the local knowledge base is used to strictly limit the answer of the large model (which is explicitly marked in the prompt word that the answer is strictly performed according to the given content without jumping out of the given text range), so as to ensure that the answer of the AI does not exceed the range of the part-time industry. In addition, after the answer result is obtained, content security detection is performed on the result to further guarantee the security of the output result of the large model.
In this embodiment, as shown in fig. 2, the capturing the intent of the user accurately by using the AI large model based on the information input by the user includes:
and performing pre-layer processing on characters or voices input by a user, converting the voices into characters, detecting and intercepting sensitive words, correcting wrongly written characters and/or removing special characters in a translation mode, then assembling the processed input information into prompt words, and then transmitting the assembled prompt words into an AI large model for intention recognition and obtaining results. For example, a simple hint word example is: "you are a part job domain senior assistant, knowing that the main intention of the user's question is: 1. finding out work; 2. inquiring post information; 3. inquiring about the interview arrangement; 4. other intents. Now, you need to judge what kind of user's intention is according to the user's input information: i want to find work. Based on the powerful capabilities of the existing AI large model, after this layer, the real intent of the user can be basically obtained.
As an infinite number of intents are involved in the actual situation, several intents that are important in the actual business are listed in this embodiment, including work intention, inquiry post information intention, inquiry interview arrangement intention, and others.
(1) Aiming at the searching work intention, the invention is a dialogue type part-time recruitment method, so that the occupation of the user triggering the searching work intention is more. Thus, after the find intent is triggered, the intelligent find process workflow is entered.
Because the part-time workers have certain requirements on the age, sex and academic of the users. Firstly, judging whether to guide the user to perfect resume information according to the resume integrity of the current user, wherein the perfect resume information is completed by a dialogue flow. For example, the user can carry out a real post recommendation process after the age, sex and academic are all perfect, three slots are set to store age, sex and academic information, and intelligent recognition is carried out according to user history information and dialogue context information to fill the three slots. When a slot, such as an academic miss, is interrogated using a procedure similar to the following: "please ask what is your academic? Of course, there may be better ways of interaction, such as showing "doctor, master, university, family, specialty, high school.
After the basic information required by the recommended post is collected, entering a link for searching and recommending the post by using a local knowledge base. In the link, knowledge enhancement is performed on the session information of the current user, for example, the user inputs "part-time role near the scene creation path which I want to find", here, "creation Jing Lu" is recognized and extracted through a named entity, the scene creation path is input into a knowledge graph, subway and business district information such as subway No. 5 lines, wandang squares and the like can be extracted, and the information is structured through the knowledge graph and then is transmitted into a local knowledge base so that relevant posts can be quickly obtained for recommendation. The recommended posts can achieve better interactive experience effect after being displayed according to a certain display template in a chat reply mode. In addition, if the user's input is "do there is a post similar to this post? When the method is used, the AI large model can identify that the user intends to recommend the intention for the similar positions, and the user can process the information of the current position by using the method and then recommend the information to the position matched with the similarity top in the vector database.
(2) Aiming at the purpose of inquiring post information, the industrial bit stream firstly combines dialogue context to take out user input, then combines recommended post information to complete prompt word assembly, and then transmits the assembled prompt word into an AI large model, and the AI large model calls a local knowledge base to answer, and meanwhile carries out security detection on answer content. Specifically, if the user wants to ask questions such as welfare treatment, notes, etc. after obtaining the recommended post, the AI large model can answer according to the local database. For example, when the question of the user is "how good is at this post", the following prompt word assembly is performed, "you are a recruiter with a deep job, and the information of this post is: .. you can only answer the recruiter's question based on the above post information: how well the welfare treatment of this post is. The assembled prompt words are transmitted into the large model and are answered in combination with the contents in the local database. The content of the answer also needs to be subjected to content security detection, so that the situations of messy answer, answer beyond the industry range and the like are prevented.
(3) Aiming at inquiring interview arrangement intention, the industrial bit stream is used for completing the assembly of the prompt word according to interview time filled in when the merchant issues posts in a local knowledge base, then the assembled prompt word is transmitted into an AI large model, the AI large model answers the user, meanwhile, safety detection is carried out on the answer content, after the job seeker confirms, the answer content is sent to the merchant in a short message or other channel mode, after the merchant confirms secondarily, the prompt is returned, otherwise, the job seeker or the merchant waits for confirmation again. Specifically, in the previous recruitment process, the process of scheduling interviews is complex because of the need of connecting employers and job seekers in series, and the situations that both parties are not online at the same time are relatively many. Thus, due to the untimely synchronization of information, there is no way for employers to create a good use experience. According to the invention, according to the proper interview time recorded in a local database by an employer, namely a merchant, then when questions and answers are received, the interview time is arranged by assembling a prompt word and transmitting the prompt word into a large model, and then the interview time is confirmed by a job seeker and the employer, any party confirms that the interview is not passed, and a round of arrangement is carried out, and more than three times of arrangement are not passed and can be manually interposed. The general prompt word is "you are a recruiter, you need to interview the job seeker according to the appropriate interview time, which is. Meanwhile, the content of the answer also needs to be subjected to content security detection, so that the situations of messy answer, answer beyond the industry range and the like are prevented.
(4) For other intents, such as question, chat, opinion feedback, and the like, the workflow is to update a local knowledge base according to feedback information and optimize prompt words and flows of the AI large model. Taking opinion feedback as an example, each answer given in the AI big model allows the user to directly give evaluation and feedback, and the merchant can also feed back the recruited service satisfaction of the present invention, which will be used to optimize the model, prompt and flow of the present invention.
The present embodiment is merely illustrative of a few of the above-mentioned relatively important intents, and in the field of part-time conversational recruitment, the user may have a variety of intents, and other such intents may be within the scope of the present invention.
In summary, the invention discloses a set of conversational part-job recruitment service method based on a local part-job information knowledge base composed of a knowledge graph of the part-job industry, an industry vector database and the like and AIGC (automatic guided vehicle) capability of a large model in the industry. The novel part-time recruitment service mode greatly shortens the part-time finding path of the user, replaces the original service flow with the intelligent workflow, simplifies the original information synchronization flow and greatly improves the part-time service efficiency.
Claims (9)
1. The conversational part-part recruitment method based on the local knowledge base and the AI large model is characterized by comprising the following steps of: the method comprises the following steps:
based on information input by a user, accurately capturing the intention of the user by using an AI large model;
customizing a workflow for each intention according to the disagreement graph of the user, and calling a local knowledge base in the workflow to assist in completing the recruitment of the part-time job.
2. The conversational, part-time recruitment method of a local knowledge base and AI big model of claim 1, wherein: the local database comprises a knowledge graph and a vector database;
the knowledge graph is formed by carrying out entity identification on part-time information, and carrying out information classification and relation association by utilizing industry classification and Embedding similarity to form the knowledge graph;
the vector database is formed by selecting a pyrtorch version of an open-source bert-base-Chinese model to perform an encoding process on the part-time text information and then storing the part-time text information in the database.
3. The conversational, part-time recruitment method of a local knowledge base and AI big model of claim 2, wherein: the entity identification shows that the user is selected to search hot words, cities, business circles, subways, schools and post category data, the data are marked in detail and then used as industry word stock, then a hanNLP tool is used for named entity identification, and the associated knowledge graph data are stored in an open-source graph database neo4 j.
4. The conversational, part-time recruitment method of a local knowledge base and AI big model of claim 1, wherein: the method for accurately capturing the intention of the user by using the AI large model based on the information input by the user comprises the following steps:
the method comprises the steps of performing pre-layer processing on characters or voices input by a user, converting the voices into characters, detecting and intercepting sensitive words, correcting wrongly written characters and/or removing special characters in translation mode, then assembling the processed input information into prompt words, and then transmitting the assembled prompt words into an AI large model for intention recognition and obtaining results.
5. The conversational, part-time recruitment method of a local knowledge base and AI big model of claim 4, wherein: the intent recognition includes finding work intent, asking post information intent, asking interview arrangement intent, and other intents.
6. The conversational, part-time recruitment method of a local knowledge base and AI big model of claim 5, wherein: aiming at finding the work intention, the workflow is to check the resume information of the user first, if the resume information is complete, the input information is subjected to entity identification by utilizing local knowledge; if the information is incomplete, guiding the user to perfect necessary filling information through multiple rounds of dialogue, and then carrying out entity identification on the input information by utilizing local knowledge; and then searching and recommending the information from the local knowledge base, and finally outputting recommended post information according to a template format.
7. The conversational, part-time recruitment method of a local knowledge base and AI big model of claim 5, wherein: aiming at the purpose of inquiring post information, the industrial bit stream firstly combines dialogue context to take out user input, then combines recommended post information to complete prompt word assembly, and then transmits the assembled prompt word into an AI large model, and the AI large model calls a local knowledge base to answer, and meanwhile carries out security detection on answer content.
8. The conversational, part-time recruitment method of a local knowledge base and AI big model of claim 5, wherein: aiming at inquiring interview arrangement intention, the industrial bit stream is used for completing the assembly of the prompt word according to interview time filled in when the merchant issues posts in a local knowledge base, then the assembled prompt word is transmitted into an AI large model, the AI large model answers the user, meanwhile, safety detection is carried out on the answer content, after the job seeker confirms, the answer content is sent to the merchant in a short message or other channel mode, after the merchant confirms secondarily, the prompt is returned, otherwise, the job seeker or the merchant waits for confirmation again.
9. The conversational, part-time recruitment method of a local knowledge base and AI big model of claim 5, wherein: for other purposes, the workflow updates the local knowledge base according to the feedback information, and optimizes the prompting words and the flow of the AI large model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311178181.2A CN117474505A (en) | 2023-09-13 | 2023-09-13 | Conversational part-part recruitment method based on local knowledge base and AI large model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311178181.2A CN117474505A (en) | 2023-09-13 | 2023-09-13 | Conversational part-part recruitment method based on local knowledge base and AI large model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117474505A true CN117474505A (en) | 2024-01-30 |
Family
ID=89624552
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311178181.2A Pending CN117474505A (en) | 2023-09-13 | 2023-09-13 | Conversational part-part recruitment method based on local knowledge base and AI large model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117474505A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118037250A (en) * | 2024-04-11 | 2024-05-14 | 成都鱼泡科技有限公司 | Data mining method and system applying text informatization system |
-
2023
- 2023-09-13 CN CN202311178181.2A patent/CN117474505A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118037250A (en) * | 2024-04-11 | 2024-05-14 | 成都鱼泡科技有限公司 | Data mining method and system applying text informatization system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6793975B2 (en) | Video-based Jobs Job Matching Servers and methods and computer-readable recording media containing programs to perform those methods | |
| CN110266899B (en) | Client intention identification method and customer service system | |
| CN107153965A (en) | A kind of intelligent customer service solution of multiple terminals | |
| CN110825858A (en) | Intelligent interaction robot system applied to customer service center | |
| CN110459210A (en) | Answering method, device, equipment and storage medium based on speech analysis | |
| CN109688281A (en) | A kind of intelligent sound exchange method and system | |
| CN105744057A (en) | Intelligent voice dialogue interaction method and intelligent voice dialogue interaction device | |
| CN111241260A (en) | Data processing method, device and equipment based on human-computer interaction and storage medium | |
| CN112925888B (en) | Training method and device for question-answer response and small sample text matching model | |
| CN117573834B (en) | Multi-robot dialogue method and system for software-oriented instant service platform | |
| CN116595148B (en) | Method and system for realizing dialogue flow by using large language model | |
| CN111651554A (en) | Insurance question-answer method and device based on natural language understanding and processing | |
| CN114817507A (en) | Reply recommendation method, device, equipment and storage medium based on intention recognition | |
| CN110321414A (en) | A kind of artificial intelligence counseling services method and system based on deep learning | |
| CN112860873A (en) | Intelligent response method, device and storage medium | |
| CN117094690A (en) | Information processing method, electronic device, and storage medium | |
| CN117370512A (en) | Method, device, equipment and storage medium for replying to dialogue | |
| CN117474505A (en) | Conversational part-part recruitment method based on local knowledge base and AI large model | |
| CN116108141A (en) | Similar question searching mechanism under interview scene | |
| CN114004599A (en) | Material demand plan examination system based on artificial intelligence | |
| WO2020124444A1 (en) | Information processing method and related apparatus | |
| CN116578682B (en) | Intelligent consultation method and system for legal service | |
| CN109146725A (en) | It is a kind of for support robot provide legal advice service cloud platform | |
| CN114117024B (en) | Platform construction method for multi-round conversation function scene | |
| US20240143643A1 (en) | Dynamic prediction of agenda item coverage in a communication session |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20240218 Address after: 311100 room 1306, jinzhiyuan building, Cangqian street, Yuhang District, Hangzhou City, Zhejiang Province Applicant after: Hangzhou qingtuanbao Network Technology Co.,Ltd. Country or region after: China Address before: 11 / F, jinzhiyuan building, Wenyi West Road, Yuhang District, Hangzhou, Zhejiang 310000 Applicant before: HANGZHOU HUTU TECHNOLOGY Co.,Ltd. Country or region before: China |