An intelligent question-answering method and system

Abstract

The application belongs to the technical field of natural language processing, and provides an intelligent question and answer method and system. The method comprises the following steps: extracting key word information in a to-be-processed question and answer topic, matching the key word information with a database to generate a matching success signal or a matching failure signal; for the matching failure signal, performing secondary recognition analysis on the to-be-processed question and answer topic, further performing segmentation to obtain a topic segmentation segment, and matching the topic segmentation segment with the database to generate a matching result; the secondary recognition analysis on the to-be-processed question and answer topic comprises determining a segmentation point, performing topic segmentation, analyzing the multiple segmentation segments obtained by segmentation, and performing sorting calculation, simultaneously generating display information according to a priority sorting value, and transmitting the display information to a corresponding user. The application can perform secondary recognition analysis on the question and answer topic, further perform secondary matching on the multiple segmentation segments, and obtain more optimal display information.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of natural language processing, and particularly relates to an intelligent question and answer method and system. BACKGROUND

[0002] A question and answer system is a high-level form of an information retrieval system, which can answer a user's natural language question with accurate and concise natural language, so as to meet the demand of people for quickly and accurately obtaining information. With the development of the field of artificial intelligence and natural language processing, the question and answer system is more and more concerned by people.

[0003] In the prior art (application No. CN201510976272.X), an intelligent question and answer system is disclosed, which comprises: one or more question input modules, the question input module is used to input a question to be answered, and sets a source identifier for the question, the source identifier contains at least two or more large user information, the large user further includes a small user, the large users are independent of each other, and the small users of different large users are independent; one question library; one question matching module; one general answer library; one or more professional answer libraries; and one question answering module, which finds an answer of a corresponding question in the professional answer library or the general answer library according to a matching result of the question matching module and the source identifier of the question, and feeds back the corresponding answer to an answer requiring party according to the source identifier. However, when some existing intelligent question and answer systems are used, the system cannot recognize the question input by the user when the user inputs the question to interact and ask questions, which further causes the question and answer system to fail to achieve the purpose of asking and answering. In addition, when the question and answer result is displayed, a large amount of data is not comprehensively sorted, the question and answer result cannot be displayed according to the characteristics of the question and answer result, and the user experience is reduced.

[0004] Therefore, it is necessary to provide an intelligent question and answer method to solve the above problems. SUMMARY

[0005] The present application aims to provide an intelligent question and answer method and system to solve the technical problems that the existing method in the prior art cannot recognize the question input by the user, cannot obtain a reply, and cannot comprehensively sort data according to the characteristics of the data to obtain a more effective reply. The technical problems to be solved by the present application are solved by the following technical solutions.

[0006] The first aspect of the present application provides an intelligent question and answer method, comprising: analyzing an obtained question and answer question to be processed, extracting key word information in the question and answer question to be processed, and matching the key word information with a database to generate a matching success signal or a matching failure signal; for the matching failure signal, performing secondary recognition analysis on the question and answer question to be processed, segmenting the question and answer question to be processed according to the length of the key word information to obtain a question segmentation segment, and matching the question segmentation segment with the database to generate a matching result; the secondary recognition analysis on the question and answer question to be processed comprises comparing the length of common information in the question and answer question to be processed with the length of the key word information to determine a segmentation point for question segmentation, and analyzing the plurality of segmentation segments obtained by segmentation; analyzing the generated matching result, screening to obtain similar results and non-similar results, and sorting and calculating the similar results and non-similar results, simultaneously generating display information according to a priority sorting value, and transmitting the display information to a corresponding user.

[0007] The second aspect of the present application provides an intelligent question and answer system for executing the intelligent question and answer method of the first aspect of the present application, comprising: a first analysis model for analyzing an obtained question and answer question to be processed, extracting key word information in the question and answer question to be processed, and matching the key word information with a database to generate a matching success signal or a matching failure signal; a second analysis module for, for the matching failure signal, performing secondary recognition analysis on the question and answer question to be processed, segmenting the question and answer question to be processed according to the length of the key word information to obtain a question segmentation segment, and matching the question segmentation segment with the database to generate a matching result; the secondary recognition analysis on the question and answer question to be processed comprises comparing the length of common information in the question and answer question to be processed with the length of the key word information to determine a segmentation point for question segmentation, and analyzing the plurality of segmentation segments obtained by segmentation; a calculation module for analyzing the generated matching result, screening to obtain similar results and non-similar results, and sorting and calculating the similar results and non-similar results, simultaneously generating display information according to a priority sorting value, and transmitting the display information to a corresponding user.

[0008] The embodiments of the present application have the following advantages:

[0009] Compared with the prior art, the present application can perform secondary recognition analysis on the question and answer question for the case that cannot be recognized, especially the case that one-time recognition cannot obtain a recognition result, by determining a segmentation point and segmenting according to different information in the question and answer question in different ways, and further performing secondary matching on the plurality of segmentation segments, and comprehensively arranging the matching results according to their own characteristics, so that the matching results can be more effectively displayed according to their characteristics, the results with high matching degree are preferentially displayed, and the user can conveniently view. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating an example of the intelligent question-answering method of the present invention;

[0011] Figure 2 This is a flowchart illustrating one embodiment of the intelligent question-answering method of the present invention;

[0012] Figure 3 This is a flowchart illustrating the process of searching and matching question segments in the database in the question-answering method of the present invention;

[0013] Figure 4 This is a schematic diagram of an example of a question-and-answer system according to the present invention;

[0014] Figure 5 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention;

[0015] Figure 6 This is a schematic diagram of a computer-readable medium embodiment according to the present invention. Detailed Implementation

[0016] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0017] In view of the above problems, this invention proposes an intelligent question-answering method. This method analyzes the acquired question-answering questions, extracts keyword information from the questions, and matches the keyword information with a database to generate a successful match signal or a failed match signal. For the failed match signal, a secondary identification analysis is performed on the question-answering questions to segment them according to the length of the keyword information, and the segmented questions are then matched with the database to generate matching results. The secondary identification analysis includes comparing the length of commonly used information and the length of keyword information in the question-answering questions to determine the segmentation point for question segmentation, and analyzing the multiple segmented segments. The generated matching results are analyzed, similar and dissimilar results are filtered out, and the similar and dissimilar results are sorted and calculated. Display information is generated according to the priority ranking value and transmitted to the corresponding user.

[0018] The following reference Figures 1 to 3 The present invention will be described in detail below.

[0019] Figure 1 This is a flowchart illustrating an example of the intelligent question-answering method of the present invention.

[0020] Next, in step S101, the acquired questions to be processed are analyzed. Keyword information in the questions to be processed is extracted and matched with the database to generate a matching success signal or a matching failure signal.

[0021] In one specific implementation, the process involves acquiring the question-and-answer topic to be processed (e.g., "What is today's temperature?", or related to the interaction process of medical remote operation data, medical consultation, shopping Q&A, and smart city construction Q&A). Simultaneously, keywords within the question-and-answer topic are identified. Specifically, frequently used information from the question-and-answer topic is acquired, where frequently used information refers to common characters or words. This frequently used information is then removed from the question-and-answer topic, and the remaining information is marked as keyword information. Next, the keyword information is matched against a database. This database stores data related to the interaction process of medical remote operation data, shopping Q&A, smart city construction Q&A, weather Q&A, etc.

[0022] In one specific implementation, for example in the medical field, in a municipal hospital, when a user is seeking medical advice, they can use a self-service machine in the lobby for automated inquiry processing. For instance, if the user's question is "How to go to the registration desk to register?", the self-service machine will analyze the question, identify the text topic, extract the corresponding keywords "registration desk" and "registration", and search and match them in the database to obtain information related to the keywords. The search will be performed according to the order in which the keywords appear, prioritizing the search for "registration desk" and then performing related searches for "registration".

[0023] Based on the search results, a corresponding route is generated according to the location of the "registration desk" and displayed to the relevant personnel using voice prompts.

[0024] Specifically, the following steps are used to extract keyword information corresponding to various application scenarios (such as the interaction process of medical remote operation data, medical consultation, shopping Q&A, and smart city construction Q&A):

[0025] Step S201: Define the text topics for each application scenario (specifically including the interaction process of medical remote operation data, medical consultation, shopping Q&A, and smart city construction Q&A).

[0026] Before extracting keywords, it is necessary to clarify the text theme of the question and answer to be processed, that is, to understand the core issue, main object or key event of the text.

[0027] Step S202: Analyze the text structure of the questions to be processed.

[0028] Specifically, the title and subtitle of the question and answer topic to be processed are analyzed. The title is usually a core summary of the text and contains important keywords. The subtitle is usually a text that further explains the topic and can also provide more keyword clues. The beginning and end of a text paragraph contain the key information of the text paragraph. Key sentences: key sentences in the text, such as conclusion sentences, theme sentences, etc., usually contain important keywords.

[0029] Step S203: Analyze the nouns and verbs in the question and answer topic to be processed.

[0030] The nouns and verbs in the question and answer topic to be processed are analyzed and determined to correspond to each application scenario (for example, the interaction process of medical remote operation data, medical diagnosis, shopping question and answer, and city intelligent construction question and answer).

[0031] In addition, by counting the high-frequency words in the text of the question and answer topic to be processed, the keywords are included in these words.

[0032] Step S204: Verify the extracted keywords.

[0033] After extracting the keywords, the extracted keywords are verified. The keywords are combined to see if they can accurately summarize the main content of the text. If the keywords cannot well reflect the core of the text, the keywords need to be re-extracted or adjusted.

[0034] For example, for the following text: “With the continuous development of artificial intelligence technology, machine learning algorithms have achieved remarkable results in image recognition, natural language processing, etc. Deep learning, as an important branch of machine learning, has realized efficient processing of professional data in city intelligent construction question and answer through the construction of deep neural networks. The professional data includes road length, road position coordinates, building height, building area, etc.”

[0035] Specifically, the keywords are extracted according to the above steps S201-S204. The extraction and analysis results include the nouns of the above text, including “artificial intelligence”, “machine learning”, “deep learning”, “city intelligent construction question and answer”, and “professional data”. These are keywords. Among them, “artificial intelligence”, “machine learning”, and “deep learning” are all professional terms. “Professional data” appears twice and has a high frequency in the above text, so it is also a keyword. The following nouns in the text are further explanations of the keyword “professional data”: road length, road position coordinates, building height, and building area. Therefore, they are also extracted and used as keywords.

[0036] In another specific embodiment, the question-answer question to be processed is "What are the examination items for a certain disease in the medical field?" First, it is determined that the applicable application direction is medical. Then, the commonly used information in the question-answer to be processed, such as "in", "of", "what are", and the like, is analyzed, and after the commonly used information is removed, the remaining "medical field", "certain disease", and "examination items" are marked as keyword information, and then matched with the above database. When the database stores various medical field disease information, examination item data, and related research results and the like, the corresponding information is extracted as the answer text to the question-answer question to be processed.

[0037] When the keyword information is successfully matched with the keyword and / or text segment in the database, a matching success signal is generated, and the matching result corresponding to the matching success is acquired and transmitted to the matching result analysis module. On the contrary, when the keyword information is not successfully matched with the keyword and / or text segment in the database, a matching failure signal is generated. The text segment information includes the subject name representing each application scenario, the noun and the limiting word appearing more than a set number of times (for example, twice, three times, or more).

[0038] It should be noted that the above is only described as an optional example and cannot be understood as a limitation of the present application.

[0039] Next, in step S102, for the matching failure signal, the question-answer question to be processed is analyzed twice to determine the key information and the commonly used information, to segment the question, and analyze the multiple segments obtained by segmentation.

[0040] For the matching failure signal, the question-answer question to be processed (for example, what question-answer question, combined with an example of an application scenario) is analyzed twice (corresponding to "second matching" in Figure 2 ), to determine the key information and the commonly used information, to segment the question (corresponding to "information segmentation" in Figure 2 ), and analyze the multiple segments obtained by segmentation, to determine the matching result of the second matching.

[0041] When the length of the commonly used information in the question-answer question to be processed is greater than the length of the key information, the commonly used information is used as the segmentation point, and the question-answer question to be processed is segmented to generate multiple question segmentation segments.

[0042] When the length of the common information in the question-answer question to be processed is less than or equal to the length of the keyword information, the keyword information is taken as the segmentation point, and the question-answer question to be processed is segmented to generate multiple question segmentation segments.

[0043] The question-answer question to be processed is segmented according to the length of the keyword information to obtain question segmentation segments, and the question segmentation segments are matched with the keywords and / or text segments in the database to generate matching results. The text segment information includes the topic name representing each application scenario, and the noun and the determiner appearing more than a set number of times (for example, twice, three times, or more).

[0044] Specifically, the number of common information and keyword information in the question-answer question to be processed is recorded, and is recorded as the length L1 of the common information and the length L2 of the keyword information (specifically, the total number of words of the common information and the keyword information, also referred to as the length L1 of the common information and the length L2 of the keyword information), and the length L1 of the common information and the length L2 of the keyword information are compared. When the length L1 of the common information is greater than the length L2 of the keyword information, the common information is taken as the segmentation point to segment the question-answer question to be processed to generate question segmentation segments. When the length L1 of the common information is less than the length L2 of the keyword information, the keyword information is taken as the segmentation point to segment the question-answer question to be processed to generate question segmentation segments.

[0045] In a specific embodiment, the question-answer question to be processed is “What is the temperature of the weather today?”, the question-answer question to be processed as a whole is matched with the database, if there is a corresponding matching result, the corresponding matching result (for example, 25-30 degrees Celsius) is directly obtained.

[0046] If it is a non-matching result, the question segmentation segments generated by segmentation are matched again, for example, segmented into “today”, “weather temperature”, “is”, and “how much”.

[0047] Specifically, as shown in the above example, the keyword information is “today” and “weather temperature”, and the corresponding common information is “how much” and “is”.

[0048] As shown in Figure 3 The following steps are performed to analyze the multiple segmentation segments obtained by segmentation:

[0049] Step S301: The question-answer question to be processed is matched with the database, if there is a corresponding matching result, the matching result is filtered, and the question segmentation segments generated are matched with the database again.

[0050] Step S302: the generated question segmentation is labeled as i, and i=1, 2, …, j, and the length L1i of the common information and the length L2i of the keyword information in each question segmentation are obtained, and the proportion of the number of common information in the question segmentation i, i.e. the first proportion Z1i, is calculated, and the average proportion of the keyword information in all question segmentations of the to-be-processed question and answer question, i.e. the average proportion Z2p, is calculated.

[0051] Step S303: the question segmentation corresponding to the first proportion Z1i greater than the average proportion Z2p is removed, the remaining question segmentations are matched with the database, it is judged whether there is a corresponding matching result, when there is a matching result, the matching result is obtained, and the obtained matching result is transmitted to the matching result analysis module to generate matching success information; when there is no matching result, an unmatched success result is obtained, and the unmatched success result is transmitted to the matching result analysis module to generate an unmatched success signal.

[0052] The proportion of the length of the common information in the question segmentation is specifically calculated, for example, the number of question segmentations is 4, and the length of the common information is 2, so the calculated proportion is 1 / 2, and the proportion of the length of the common information in all question segmentations is calculated in the same way, and the average proportion of the keyword information in all question segmentations of the to-be-processed question and answer question, i.e. the average proportion, is calculated, and the average proportion is compared with the proportion of the length of the common information and is filtered, and the proportion of the length of the common information lower than the average value is filtered.

[0053] It should be noted that the above is only an optional example for description, and cannot be understood as a limitation of the present application.

[0054] Next, in step S103, the generated matching result is analyzed, similar results and non-similar results are filtered, and the similar results and non-similar results are sorted and calculated, and display information is generated according to the priority sorting value, and the display information is transmitted to the corresponding user

[0055] In a specific embodiment, the generated matching result is obtained, and the matching result is labeled as n, and n=1, 2, …, m, then the matching result is screened, the matching result with the same information as the to-be-processed question and answer question is recorded as a similar result a, and a=1, 2, …, o, then the remaining matching result is recorded as a non-similar result b, and b=1, 2, …, p.

[0056] The number of the same words in the similar result a and the to-be-processed question and answer question is obtained and recorded as the same word number La, and the ratio of the same word number La to the similar result a is calculated and recorded as .

[0057] The display times Ya of the similar result a in the historical display data of the specified historical time period are calculated, and the sorting order Yaq corresponding to the display times Ya is obtained, and the following expression is used to calculate the priority sorting value Ka corresponding to the similar result a:

[0058]

[0059] Wherein, Ka represents the priority sorting value corresponding to the similar result a; La represents the same number of words in the similar result a and the question and answer to be processed; Ya represents the display times of the similar result a in the historical display data of the specified historical time period; Yaq represents the sorting order corresponding to the qth display times Ya, and q=1, 2, …, u.

[0060] Specifically, the historical specified time period includes one month, two months, three months or half a year.

[0061] The display times are specifically the display times of the same display platform in the historical specified time period.

[0062] In another specific embodiment, all non-similar results b are obtained, and the non-similar results b are classified according to the similar proportion to obtain classified results b g , and g=1, 2, …, e, the proportion value of the same type of classified results in the same type of classified results is calculated, and the average proportion value of the same type of classified results is calculated and recorded as , and the average proportion value of all classified results is calculated in turn.

[0063] The display times Yb of the non-similar result b in the historical display data of the specified historical time period are calculated g , and the sorting order Yb g corresponding to the display times Yb g is obtained, q, and q=1, 2, …, u; the following expression is used to calculate the priority sorting value Kb corresponding to the non-similar result b:

[0064]

[0065] Wherein, Kb represents the priority sorting value corresponding to the non-similar result b; Yb g represents the display times of the non-similar result b in the historical display data of the specified historical time period; Yb g q represents the sorting order corresponding to the qth display times Yb g , and q=1, 2, …, u; represents the average proportion value of the same type of classified results.

[0066] Specifically, the historical specified time period includes one month, two months, three months or half a year.

[0067] The display times are specifically the display times of the same display platform in a historical specified time period.

[0068] It should be noted that the above is only described as an optional example, and cannot be understood as a limitation of the present application. In addition, the above figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present application, and are not for limiting purposes. It is easy to understand that the processes shown in the above figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be executed synchronously or asynchronously, for example, in multiple modules.

[0069] Compared with the prior art, the present application can perform secondary identification analysis on the question and answer questions by identifying and analyzing the obtained question and answer questions, and can perform secondary identification analysis on the question and answer questions in the case of being unable to identify, especially in the case of being unable to obtain identification results by one-time identification. Specifically, by determining a segmentation point and segmenting in different ways according to different information in the question and answer questions, and further performing secondary matching on multiple segmented segments, the matching results are comprehensively arranged according to their own characteristics, so that the matching results can be more effectively displayed according to their characteristics, the results with high matching degree are preferentially displayed, and the user can conveniently view.

[0070] The following is an embodiment of the system of the present application, which can be used to execute the method embodiment of the present application. For details not disclosed in the system embodiment of the present application, please refer to the method embodiment of the present application.

[0071] Figure 4 is a structural schematic diagram of an example of the intelligent question and answer system according to the present application.

[0072] As Figure 4 shown, the second aspect of the present disclosure provides an intelligent question and answer system 400, which comprises a first analysis model 410, a second analysis model 420 and a calculation model 430.

[0073] In a specific embodiment, the first analysis model 410 analyzes the acquired to-be-processed question and answer question, extracts keyword information in the to-be-processed question and answer question, and matches the keyword information with the database to generate a matching success signal or a matching failure signal. The second analysis model 420, for the matching failure signal, performs secondary identification analysis on the to-be-processed question and answer question, divides the to-be-processed question and answer question according to the length of the keyword information to obtain question division segments, and matches the question division segments with the database to generate a matching result. The secondary identification analysis includes comparing the length of common information and the length of keyword information in the to-be-processed question and answer question to determine a division point for question division, and analyzing the obtained multiple division segments. The calculation model 430 analyzes the generated matching result, filters similar results and non-similar results, sorts and calculates the similar results and non-similar results, generates display information according to a priority sorting value, and transmits the display information to a corresponding user.

[0074] According to an optional embodiment, the length of common information and the length of keyword information in the to-be-processed question and answer question are compared to determine a division point for question division.

[0075] Specifically, the to-be-processed question and answer question with a matching failure signal is subjected to secondary identification to determine keyword information and common information.

[0076] When the length of common information in the to-be-processed question and answer question is greater than the length of keyword information, the common information is taken as a division point, and the to-be-processed question and answer question is divided to generate multiple question division segments. When the length of common information in the to-be-processed question and answer question is less than or equal to the length of keyword information, the keyword information is taken as a division point, and the to-be-processed question and answer question is divided to generate multiple question division segments.

[0077] According to an optional embodiment, the analysis of the obtained multiple division segments specifically includes:

[0078] S1: The to-be-processed question and answer question is matched with the database. If there is a corresponding matching result, the matching result is filtered, and the generated question division segments are matched with the database again;

[0079] S2: The generated question division segments are labeled as i, and i=1, 2, …, j. The length L1i of common information and the length L2i of keyword information in each question division segment are obtained, and the proportion of the number of common information in the question division segment i, i.e., the first proportion Z1i, is calculated. In addition, the average proportion of keyword information in all question division segments of the to-be-processed question and answer question, i.e., the average proportion Z2p, is calculated.

[0080] S3: the first proportion value Z1i is greater than the proportion mean Z2p corresponding to the question segmentation section is eliminated, the remaining question segmentation section is matched with the database, it is judged whether there is corresponding matching result, when there is matching result, the matching result is obtained, and the matching result is transmitted to the matching result analysis module, and matching success information is generated; when there is no matching result, the unmatching success result is obtained, and the unmatching success result is transmitted to the matching result analysis module, and unmatching success signal is generated.

[0081] According to an optional embodiment, the generated matching result is analyzed, similar results and non-similar results are screened, and the similar results and non-similar results are sorted and calculated.

[0082] Specifically, all matching results are obtained and the matching results are labeled as n, and n=1, 2, …, m, then the matching results are screened, the matching results with the same information as the question and answer to be processed are recorded as similar results a, and a=1, 2, …, o, then the remaining matching results are recorded as non-similar results b, and b=1, 2, …, p.

[0083] The number of same words in the similar results a and the question and answer to be processed is obtained and recorded as the same word number La, and the ratio of the same word number La to the similar results a is calculated and recorded as .

[0084] The number of times Ya of the similar results a in the historical display data of the specified historical time period is calculated, and the sorting order Yaq corresponding to the number of times Ya is obtained, and the following expression is used to calculate the priority sorting value Ka corresponding to the similar results a:

[0085]

[0086] Wherein, Ka represents the priority sorting value corresponding to the similar results a; La represents the number of same words in the similar results a and the question and answer to be processed; Ya represents the number of times of the similar results a in the historical display data of the specified historical time period; Yaq represents the sorting order corresponding to the qth display number Ya, and q=1, 2, …, u.

[0087] According to an optional embodiment, further comprising: obtaining all non-similar results b, and classifying the non-similar results b according to the similar proportion to obtain classification results b g , and g=1, 2, …, e, then the proportion value of the same type classification result and the question and answer to be processed is calculated, and the proportion mean of the same type classification result is calculated and recorded as , and the proportion mean of all classification results is calculated in turn.

[0088] Calculate the number of times of display Yb of the non-similar result b in the historical display data of the specified historical time period g At the same time, the number of times of display Yb is obtained g The corresponding sorting order Yb g q, and q=1, 2, …, u; the following expression is used to calculate the priority sorting value Kb corresponding to the non-similar result b:

[0089]

[0090] Wherein, Kb represents the priority sorting value corresponding to the non-similar result b; Yb g represents the number of times of display Yb of the non-similar result b in the historical display data of the specified historical time period; Yb g q represents the number of times of display Yb of the non-similar result b for the qth time g The corresponding sorting order, and q=1, 2, …, u; represents the average proportion of the same type of classification result.

[0091] According to an optional embodiment, the step of extracting the keyword information in the to-be-processed question and answer question includes the following steps: determining the text theme corresponding to each application scenario; analyzing the text structure of the to-be-processed question and answer question; analyzing the nouns and verbs of the to-be-processed question and answer question; and verifying the extracted keywords.

[0092] Specifically, the specified historical period includes one month, two months, three months, or half a year.

[0093] When the secondary identification analysis obtains a non-matching successful result, the non-matching successful result is output to the user.

[0094] It should be noted that in this embodiment, the intelligent question and answer method performed by the intelligent question and answer system is substantially the same as the intelligent question and answer method in Figure 1 Therefore, the same part is omitted.

[0095] Figure 5 is a structural schematic diagram of an electronic device embodiment according to the present application.

[0096] As shown in Figure 5 , the electronic device is in the form of a general computing device. The processor can be one or multiple and work cooperatively. The present application also does not exclude distributed processing, i.e., the processor can be dispersed in different entity devices. The electronic device of the present application is not limited to a single entity, but can also be the sum of multiple entity devices.

[0097] The memory stores computer executable programs, usually machine readable codes. The computer readable program can be executed by the processor to enable the electronic device to perform the method of the present application, or at least part of the steps in the method.

[0098] The memory includes volatile memory, such as random access memory (RAM) and / or cache memory, and / or non-volatile memory, such as read-only memory (ROM).

[0099] Optionally, the electronic device further comprises an I / O interface for data exchange between the electronic device and external devices. The I / O interface can be one or more of a variety of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of a variety of bus structures.

[0100] It should be understood that Figure 5 The electronic device shown is only an example of the present application, and the electronic device of the present application can also include elements or components not shown in the above examples. For example, some electronic devices also include a display unit such as a display screen, and some electronic devices also include human-computer interaction elements such as buttons and keyboards. As long as the electronic device can execute the computer-readable program in the memory to implement the method or at least part of the steps of the method of the present application, it can be considered as an electronic device covered by the present application.

[0101] From the above description of the embodiments, those skilled in the art will readily understand that the example embodiments described herein can be implemented by software or by software in combination with necessary hardware. Therefore, as Figure 6 As shown, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.) or on a network, and includes a number of commands to make a computing device (which can be a personal computer, a server, or a network device, etc.) execute the above-mentioned method according to the embodiments of the present application.

[0102] The software product can use any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: electrical connections having one or more wires, portable disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0103] The computer readable storage medium can include a non-transitory computer-readable medium (e.g., volatile or non-volatile memory device), a media such as a magnetic or optical storage media. As used herein, the term "non-transitory" merely means a computer-readable medium that does not include a transitory propagating signal per se (e.g., a propagating electromagnetic wave carrying the computer-readable instructions), but does include all tangible media that can be accessed by a computer or a device for the purposes of reading and processing the computer-readable instructions.

[0104] The program code may¬be implemented in any of one or more programming languages, including an object-oriented programming language such as Java, C++, or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may- execute entirely on the user's computing device, partly on the user's computing device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device or entirely on the remote computing device or server. In the latter scenario, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computing device, such as through the Internet using an Internet Service Provider (ISP).

[0105] The computer readable medium described above can be a non-transitory computer readable medium. The above described computer readable medium can be used to store data relating to the present disclosure, such as the one or more programs (e.g., computer executable programs) for causing the computer readable medium to implement the methods of the present disclosure when executed by the device.

[0106] Those skilled in the art can understand that the above-mentioned modules can be distributed in the device according to the description of the embodiments, and can also be changed in one or more devices different from the embodiments. The modules of the above-mentioned embodiments can be combined into one module, or can be further split into a plurality of sub-modules.

[0107] From the above description of the embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solutions according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.) or a network, and includes a plurality of commands to make a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) execute the methods according to the embodiments of the present application.

[0108] The exemplary embodiments of this application are specifically illustrated and described herein. However, it is to be understood that the application is not limited to the details of the illustrated implementations; rather, this application is intended to cover any and all modifications and equivalents within the spirit and scope of the appended claims.

Claims

1. An intelligent question answering method, characterized by, The method comprises the following steps: analyzing the obtained to-be-processed question and answer question, extracting key word information in the to-be-processed question and answer question, and matching the key word information with a database to generate a matching success signal or a matching failure signal; for the matching failure signal, performing secondary identification analysis on the to-be-processed question and answer question, segmenting the to-be-processed question and answer question according to the length of the key word information to obtain question segmentation segments, and matching the question segmentation segments with the database to generate a matching result; the secondary identification analysis comprises comparing the length of common information in the to-be-processed question and answer question with the length of the key word information to determine a segmentation point for segmenting the question, and analyzing the segmented segments; analyzing the generated matching result, screening similar results and non-similar results, sorting and calculating the similar results and non-similar results, generating display information according to a priority sorting value, and transmitting the display information to a corresponding user. 2.The intelligent question answering method of claim 1, wherein, The method comprises the following steps: performing secondary identification on the to-be-processed question and answer question with the matching failure signal to determine the key word information and the common information; when the length of the common information is greater than the length of the key word information, the common information is used as the segmentation point to segment the to-be-processed question and answer question to generate multiple question segmentation segments; when the length of the common information is less than or equal to the length of the key word information, the key word information is used as the segmentation point to segment the to-be-processed question and answer question to generate multiple question segmentation segments. 3.The intelligent question-answering method of claim 1 or 2, wherein, The method comprises the following steps: S1: matching the to-be-processed question and answer question with the database, screening the matching result if there is a corresponding matching result, and matching the generated question segmentation segments with the database again; S2: labeling the generated question segmentation segments as i, i=1, 2, …, j, obtaining the length L1i of the common information and the length L2i of the key word information in each question segmentation segment, calculating a first proportion value Z1i of the common information in the question segmentation segment i, and calculating a proportion average value Z2p of the key word information in all question segmentation segments of the to-be-processed question and answer question; S3: removing the question segmentation segment with the first proportion value Z1i greater than the proportion average value Z2p, matching the remaining question segmentation segments with the database, determining whether there is a corresponding matching result, obtaining the matching result when there is a matching result, transmitting the obtained matching result to a matching result analysis module, and generating matching success information; obtaining an unmatching success result when there is no matching result, transmitting the unmatching success result to the matching result analysis module, and generating an unmatching success signal.

4. The intelligent question and answer method of claim 1, wherein when the secondary identification analysis obtains an unmatching success result, the unmatching success result is output to the user.

5. An intelligent question answering system characterized by, The intelligent question answering system comprises: A first analysis model, which analyzes the obtained question and answer to be processed, extracts key word information in the question and answer to be processed, and matches the key word information with a database to generate a matching success signal or a matching failure signal; A second analysis model, which, for the matching failure signal, performs secondary recognition analysis on the question and answer to be processed, divides the question and answer to be processed according to the length of the key word information to obtain question and answer segments, matches the question and answer segments with the database to generate a matching result, compares the length of common information with the length of the key word information in the question and answer to be processed to determine a division point for dividing the question and answer to be processed, and analyzes the multiple segments obtained by division; A calculation model, which analyzes the generated matching result, filters similar results and non-similar results, sorts and calculates the similar results and non-similar results, generates display information according to a priority sorting value, and transmits the display information to a corresponding user. 6.The intelligent question answering system of claim 5, wherein, Further comprising: performing secondary recognition on the question and answer to be processed for the matching failure signal to determine key word information and common information; when the length of the common information is greater than the length of the key word information, taking the common information as a division point to divide the question and answer to be processed to generate multiple question and answer segments; when the length of the common information is less than or equal to the length of the key word information, taking the key word information as a division point to divide the question and answer to be processed to generate multiple question and answer segments.

Images