Intelligent question answering method, device and equipment based on dual knowledge base collaboration

Abstract

The present invention discloses an intelligent question-answering method, apparatus, and device based on dual knowledge base collaboration. The method comprises: obtaining a target user's target question-answering requirement, determining whether the target question-answering requirement is of a first preset requirement type; if so, matching a target key knowledge point set corresponding to the target question-answering requirement; traversing the first target key knowledge point in the target key knowledge point set in a pre-constructed dual knowledge base to obtain a first answer case list, wherein the dual knowledge base is composed of a historical case knowledge base and a real-time case knowledge base, wherein the historical case knowledge base is constructed from screened historical typical cases; and displaying the visualized first answer case list on the target display of the target user. The present invention can effectively improve the depth of responses, enabling the target user to obtain the response they desire.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an intelligent question-answering method, device and equipment based on dual knowledge base collaboration. Background Art

[0002] With the development of information technology and artificial intelligence, intelligent question answering has been widely used. It can solve the problems of low answering efficiency and poor user experience in traditional manual question answering.

[0003] However, existing intelligent question-answering methods simply answer questions based on keyword matching, which lacks depth in responses, preventing users from obtaining the answers they expect and resulting in a poor user experience.

[0004] To address the above issues, no effective solutions have been proposed so far. Summary of the Invention

[0005] The embodiments of this specification provide an intelligent question-answering method, apparatus, and device based on dual knowledge base collaboration to solve the problem that existing intelligent question-answering methods lack depth of response.

[0006] In a first aspect, the embodiments of this specification provide an intelligent question answering method based on dual knowledge base collaboration, including:

[0007] Obtaining a target question-answering requirement of a target user, and determining whether a type of the target question-answering requirement is a first preset requirement type;

[0008] If yes, match the target key knowledge point set corresponding to the target question answering requirement;

[0009] Traversing the first target key knowledge point in the target key knowledge point set in a pre-built dual knowledge base to obtain a first answer case list, wherein the dual knowledge base is composed of a historical case knowledge base and a real-time case knowledge base, and the historical case knowledge base is constructed by screening historical typical cases;

[0010] The visualized first answer case list is displayed on a target display of a target user.

[0011] In some embodiments, determining whether the type of the target question answering requirement is a first preset requirement type includes:

[0012] Extract target keywords from the target question text corresponding to the target question answering requirement;

[0013] Determine whether the target keyword matches a first professional word identifier corresponding to the first course; if so, determine that the type of the target question-answering demand is a first preset demand type, the first course is any course in a pre-acquired target user course list, and the type of the target question-answering demand being the first preset demand type indicates that the target question-answering demand is related to the first course;

[0014] Accordingly, the target key knowledge point set corresponding to the target question answering requirement is matched, including:

[0015] Match the target key knowledge point set corresponding to the first course related to the target question answering requirement.

[0016] In some embodiments, the historical typical cases are screened in the following manner:

[0017] Extracting the first historical case record corresponding to the first case in the historical key knowledge point case record;

[0018] Reading preset case features, and collecting features of the first historical case record based on the preset case features to obtain a first feature data set;

[0019] Using the first feature data set as a clustering constraint, clustering a set of historical cases in the historical key knowledge point records to obtain a first cluster, where the set of historical cases consists of a plurality of first cases;

[0020] determining a first universality index of the first case according to a ratio of the number of cases in the first cluster to the total number of cases in the historical case set;

[0021] Reading a preset discussion feature, and collecting features of the first historical case record based on the preset discussion feature to obtain a second feature data set;

[0022] performing standardized coefficient of variation weighting processing on the second feature data set to obtain a first revealing index of the first case;

[0023] performing weighted processing on the first universality index and the first revelatory index to obtain a first typicality index of the first case;

[0024] When the first typical index reaches a preset typical threshold, the first case is regarded as a historical typical case.

[0025] In some embodiments, the historical case knowledge base is constructed from screened historical typical cases, including:

[0026] Using the first feature data set as a first label for the first case, and obtaining a key knowledge point label corresponding to the first case as a second label;

[0027] A historical case knowledge base is constructed based on the first case, the first label, and the second label.

[0028] In some embodiments, the real-time case knowledge base is constructed in the following manner:

[0029] Based on the preset case features, features of the second case are collected to obtain a third feature data set, where the second case is a case at the first time in the real-time key knowledge point case;

[0030] Using the third feature data set as a third label for the second case, and obtaining a key knowledge point label corresponding to the second case as a fourth label;

[0031] A real-time case knowledge base is constructed based on the second case, the third label, and the fourth label.

[0032] In some embodiments, traversing the first target key knowledge point in the target key knowledge point set in a pre-built dual knowledge base to obtain a first answer case list includes:

[0033] Traversing the historical case knowledge base in the pre-built dual knowledge base for the first target key knowledge point to obtain a first historical case set;

[0034] Traversing the real-time case knowledge base in the pre-built dual knowledge base for the first target key knowledge point to obtain a first real-time case set;

[0035] Combining the first historical case set with the first real-time case set to obtain a first candidate case set for a first target key knowledge point;

[0036] Calculate the matching degree between the first candidate case in the first candidate case set and the first target key knowledge point;

[0037] According to the matching degree calculation result, the first candidate cases are sorted, and target candidate cases are screened from the sorting results as the first answer case list, where the sorting threshold of the target candidate cases is greater than the preset sorting threshold.

[0038] In some embodiments, the method further comprises:

[0039] If not, the extracted target keywords are traversed in the dual knowledge base to obtain a second answer case list;

[0040] The visualized second answer case list is displayed on a target display of a target user.

[0041] In a second aspect, the embodiments of this specification further provide an intelligent question-answering device based on dual knowledge base collaboration, including:

[0042] A judgment module, configured to obtain a target question-answering requirement of a target user and judge whether the type of the target question-answering requirement is a first preset requirement type;

[0043] A matching module, configured to match a target key knowledge point set corresponding to the target question answering requirement;

[0044] a traversal module, configured to traverse a first target key knowledge point in the target key knowledge point set in a pre-built dual knowledge base to obtain a first answer case list, wherein the dual knowledge base is composed of a historical case knowledge base and a real-time case knowledge base, and the historical case knowledge base is constructed from screened historical typical cases;

[0045] The display module is used to display the first answer case list after visualization processing to the target display of the target user.

[0046] On the third aspect, an embodiment of this specification also provides an electronic device, including a memory and a processor, wherein the processor and the memory are communicatively connected to each other, the memory stores computer instructions, and the processor implements the steps of the above-mentioned intelligent question-answering method based on the collaboration of dual knowledge bases by executing the computer instructions.

[0047] In a fourth aspect, an embodiment of this specification further provides a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the steps of the above-mentioned intelligent question-answering method based on the collaboration of dual knowledge bases.

[0048] The present embodiment provides an intelligent question-answering method, apparatus, and device based on dual knowledge base collaboration. First, a target user's target question-answering requirement is obtained, and a determination is made as to whether the target question-answering requirement is of a first preset requirement type. If so, a target key knowledge point set corresponding to the target question-answering requirement is matched. Then, the first target key knowledge point in the target key knowledge point set is traversed through a pre-built dual knowledge base to obtain a first list of answer cases. The dual knowledge base comprises a historical case knowledge base and a real-time case knowledge base, wherein the historical case knowledge base is constructed from screened historical typical cases. Finally, the visualized first list of answer cases is displayed on the target user's target display. In the present embodiment, after obtaining the target question-answering requirement, a determination is made as to whether the target question-answering requirement is of the first preset requirement type. Based on the determination result, the correct answer path can be guided, effectively improving the user experience and intelligent question-answering. If the target question-answering requirement is determined to be of the first preset requirement type, a corresponding target key knowledge point set is matched, and the first target key knowledge point in the target key knowledge point set is traversed through the dual knowledge base. This effectively increases the depth of the answer, thereby enabling the target user to obtain the desired answer. The dual knowledge base consists of a historical case knowledge base and a real-time case knowledge base. The two knowledge bases can call each other, complement each other, and work together. By traversing the dual knowledge base at the first key knowledge point, a real-time, dynamic, and comprehensive first-answer case list can be obtained, solving the technical problem that the existing technology's responses lack pertinence and depth. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0050] Figure 1 This is a flow chart of an intelligent question-answering method based on dual knowledge base collaboration provided in an embodiment of this specification;

[0051] Figure 2 This is a schematic diagram of the structure of an intelligent question-answering device based on dual knowledge base collaboration provided in an embodiment of this specification;

[0052] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0053] To help those skilled in the art better understand the technical solutions in this specification, the following will provide a clear and complete description of the technical solutions in the embodiments of this specification, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. All other embodiments derived by those skilled in the art based on the embodiments in this specification without creative effort shall fall within the scope of protection of this specification.

[0054] As mentioned above, the existing intelligent question-answering methods simply answer questions based on keyword matching, which lacks the depth of the answers. As a result, target users cannot get the answers they expect, resulting in a poor user experience.

[0055] In order to solve the above problems, the embodiment of this specification provides an intelligent question-answering method based on the collaboration of dual knowledge bases. First, the target question-answering demand of the target user is obtained, and it is determined whether the type of the target question-answering demand is the first preset demand type. If so, the target key knowledge point set corresponding to the target question-answering demand is matched. Then, the first target key knowledge point in the target key knowledge point set is traversed in the pre-built dual knowledge base to obtain a first answer case list. The dual knowledge base is composed of a historical case knowledge base and a real-time case knowledge base. The historical case knowledge base is constructed by screened historical typical cases. Finally, the visualized first answer case list is displayed to the target display of the target user.

[0056] In the embodiment of this specification, after obtaining the target question-answering demand, it is first determined whether the type of the target question-answering demand is the first preset demand type. According to the judgment result, the correct question-answering path can be directed, effectively improving the user experience and the intelligence of the question-answering. After determining that it is the first preset demand type, the corresponding target key knowledge point set is matched, and the first target key knowledge point in the target key knowledge point set is traversed in the dual knowledge base, which can effectively improve the depth of the reply, so that the target user can obtain the expected reply. Among them, the dual knowledge base is composed of a historical case knowledge base and a real-time case knowledge base. The two knowledge bases can call each other, complement each other, and work together. The first key knowledge point is traversed in the dual knowledge base to obtain a real-time dynamic and comprehensive first answer case list, thereby effectively improving the user experience.

[0057] It should be noted that the terms "first," "second," and the like in the specification and claims of this application are used to distinguish similar objects, and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, for the purposes of describing the embodiments of this application.

[0058] It should also be noted that the user-related information and data involved in the embodiments of the specification of this application are all information and data authorized by the user or fully authorized by the relevant parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users or relevant parties to choose to authorize or refuse.

[0059] It is understood that the above methods provided in the embodiments of this specification can be applied to electronic devices, which can refer to electronic devices with data computing, processing, and storage capabilities. The electronic device can be a terminal such as a PC (Personal Computer), a tablet computer, a smartphone, a wearable device, an intelligent robot, etc.; it can also be a server. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0060] The following will introduce an intelligent question-answering method based on dual knowledge base collaboration provided by an embodiment of this specification in conjunction with the accompanying drawings.

[0061] Figure 1 It is a flow chart of an intelligent question-answering method based on the collaboration of dual knowledge bases provided in the embodiment of this specification. Although this specification provides the method operation steps or device structure shown in the following embodiments or drawings, the method or device may include more or fewer operation steps or module units after partial merger based on routine or no creative labor. In the steps or structures where there is no necessary causal relationship logically, the execution order of these steps or the module structure of the device is not limited to the execution order or module structure shown in the embodiments or drawings of this specification. When the method or module structure is applied to an actual device, server or terminal product, it can be executed sequentially or in parallel according to the method or module structure shown in the embodiment or drawings (for example, a parallel processor or multi-threaded processing environment, or even a distributed processing, server cluster implementation environment). For specific implementation, please refer to Figure 1 As shown, the method can be applied to an intelligent question-answering system based on dual knowledge base collaboration, and the method can include the following content.

[0062] S101: Obtain a target question-answering requirement of a target user, and determine whether the type of the target question-answering requirement is a first preset requirement type.

[0063] S102: If yes, match the target key knowledge point set corresponding to the target question answering requirement.

[0064] In the above embodiments, the target user is any user to be questioned and answered, and the target question-answering demand is realized through text, voice or other forms of interactive data input by the target user. The text or voice information input by the target user represents the needs of the target user, which is used as the target question-answering demand. The first preset demand type refers to a preset type of demand, which is usually a typical question type based on a specific field or topic. A preset classification algorithm or rule can be used to determine whether the target user's target question-answering demand is the first preset demand type, such as extracting keywords in the target question-answering demand for classification matching. This process relies on the combination of natural language processing technology, keyword matching and predetermined rules, which can accurately identify the target user's demand type and guide it to the correct question-answering path, thereby improving the intelligence level of the question-answering system and user experience.

[0065] In some embodiments, the determination of whether the target question answering requirement type is the first preset requirement type in S101 may include:

[0066] S11: extracting target keywords from the target question text corresponding to the target question answering requirement;

[0067] S12: Determine whether the target keyword matches the first professional word identifier corresponding to the first course; if so, determine that the type of the target question-answering demand is a first preset demand type, the first course is any course in the pre-acquired target user course list, and the type of the target question-answering demand being the first preset demand type indicates that the question-answering demand is related to the first course;

[0068] Accordingly, the matching of the target key knowledge point set corresponding to the target question answering requirement in S102 may include:

[0069] S13: Matching a target key knowledge point set corresponding to the first course related to the target question-answering requirement.

[0070] Specifically, the above-mentioned course list is the target major course of the target user, which may include multiple courses or course texts with key knowledge points and professional word identifiers. Among them, the key knowledge points are the core knowledge points of the course, and the professional word identifiers are specific professional field terms involved in the course content. They can be specifically set by relevant technical personnel in this field in combination with actual settings, and this manual does not make specific restrictions on this. The above-mentioned first course is any course in the course list. The first course has a corresponding first key knowledge point (first key knowledge point identifier) and a first professional word identifier. There can be more than one first key knowledge point. The process of judging whether the target question-answering demand is the first preset demand type is the above-mentioned S11-S12, which is as follows:

[0071] S11 may specifically be: using natural language processing technology to extract keywords from the target question text in the target question answering requirement as target keywords. This process includes steps such as word segmentation, entity recognition, and keyword extraction. The extracted target keywords represent the core content of the issues that the user is concerned about.

[0072] S12 may specifically be: determining whether the extracted target keyword contains the first professional word identifier, that is, whether the target keyword matches the first professional word identifier corresponding to the first course; if it matches the first professional word identifier, then the type of the target question-answering demand is considered to be the first preset demand type; at the same time, it can be determined that the target user's demand is related to the first course, and the related first course can be used as the target course. Extract the first key knowledge point in the first course (target course) related to the demand as the first target key knowledge point, and form a target key knowledge point set based on multiple first target key knowledge points. The first target key knowledge point can be any key knowledge point in the related first course.

[0073] Correspondingly, the target key knowledge point set in the above S13 includes multiple first target key knowledge points.

[0074] For example, if the target keyword of the target question text in the target user's question-answering requirement is A, and the professional term identifier of the first course is A1 under the A system, then it is considered that the target user's question-answering preset meets the first preset requirement type, and the target user's question belongs to the scope of the first course. Subsequently, the first target key knowledge point B within the relevant first course scope can be traversed in the dual knowledge base to find corresponding in-depth answer cases.

[0075] After obtaining the target user's target answer needs, by extracting the target keywords and matching them with the first professional word identifier corresponding to the target user's first course, we can combine the target user's professional background and course content to determine the courses and corresponding key knowledge points most relevant to the target question-answering needs, thereby providing users with accurate and personalized question-answering services.

[0076] In some embodiments, the above-mentioned target question text can be a structured text (data with fixed fields and types) or an unstructured text (data without fixed fields and types, such as long paragraphs of text formed by question descriptions, etc.). After the question-answering system obtains the target question-answering needs of the target user, it can first determine the type of the target text in the target question-answering needs, and then determine whether it meets the first preset demand type for different text types. For example: if the target question text is a structured text, the target question text can be directly matched with the first professional word identifier. When the matching degree meets the preset matching degree threshold, it can be determined that the type of the target question-answering demand is the first preset demand type or meets the first preset demand type; if the target question text is an unstructured text, it can be converted into a structured text first, and then matched with the first professional word identifier. Among them, the preset matching degree threshold can be set according to actual needs, and can also be determined according to the number of target keywords corresponding to the target question. This specification does not make specific restrictions on this.

[0077] S103: Traverse the first target key knowledge point in the target key knowledge point set in a pre-built dual knowledge base to obtain a first answer case list, wherein the dual knowledge base consists of a historical case knowledge base and a real-time case knowledge base, and the historical case knowledge base is constructed by screened historical typical cases.

[0078] In this embodiment, the dual knowledge base can centrally store and manage historical typical cases and real-time cases that have been screened and meet the standards. The case may include a question and the corresponding reply or answer to the question. The first reply case list may include multiple reply cases for the target question-answering needs. By traversing the first target key knowledge point in the target key knowledge point set in the pre-built dual knowledge base, the reply case related to the specific question can be quickly retrieved, facilitating intelligent question-answering and analysis.

[0079] In some embodiments, the historical typical cases in S103 above can be obtained by screening in the following manner:

[0080] S21: extracting the first historical case record corresponding to the first case in the historical key knowledge point case record;

[0081] S22: Reading preset case features, and collecting features of the first historical case record based on the preset case features to obtain a first feature data set;

[0082] S23: using the first feature data set as a clustering constraint, clustering the historical case set in the historical key knowledge point records to obtain a first cluster;

[0083] S24: determining a first universality index of the first case according to a ratio of the number of cases in the first cluster to the total number of cases in the historical case set;

[0084] S25: Reading preset discussion features, and collecting features of the first historical case record based on the preset discussion features to obtain a second feature data set;

[0085] S26: performing standardized coefficient of variation weighting processing on the second feature data set to obtain a first revelatory index of the first case;

[0086] S27: performing weighted processing on the first universality index and the first revelatory index to obtain a first typicality index of the first case;

[0087] S28: When the first typical index reaches a preset typical threshold, the first case is taken as a historical typical case.

[0088] Specifically, the above-mentioned historical key knowledge point case records refer to key knowledge point cases within the historical period, and usually contain a large number of cases related to key knowledge points. These cases include different themes, backgrounds, and processing methods. In order to effectively utilize these data, it is necessary to first conduct a multi-dimensional analysis of the historical key knowledge point case records according to a predetermined screening strategy to obtain historical typical cases, such as: screening out historical typical cases from dimensions such as the educational value of the case, the discussion heat, and the key knowledge points involved. Among them, the design of the predetermined screening strategy is usually based on several core features: the universality of the case, (educational) enlightenment, and discussion activity. Universality can be judged by analyzing indicators such as the frequency of occurrence of the case in history and the number of related discussions; educational enlightenment can be measured by evaluating the contribution of the case to education, such as whether it has triggered students' in-depth thinking and whether it has brought valuable educational experience; discussion activity reflects the social attention of the case, which is usually evaluated by analyzing the quantity and quality of social platforms, academic discussions, and media reports. Among them, the process of screening out historical typical cases is S21-S28 above, as follows:

[0089] S21 may specifically be: extracting a first historical case record corresponding to the first case from the historical key knowledge point case record, wherein the first case generally refers to any case in the historical key knowledge point case record, and the first historical case record is a historical record corresponding to the first case.

[0090] S22 may specifically be: reading relevant data of the first historical case record, collecting features of the first historical case record according to preset case features, or extracting case features in the first historical case record according to preset case features, such as: the field, object, case type, etc. of the case, to form a first feature data set.

[0091] S23 may specifically include: clustering the historical case set in the historical key knowledge point records using the first feature data set as a clustering constraint to obtain a first cluster. The purpose of cluster analysis is to group similar cases together, thereby extracting case clusters with similar characteristics. The historical case set may be a set formed based on the first case, and the first cluster may generally refer to any one of the clusters or case clusters.

[0092] S24 may specifically be: the first universality index may be determined by the ratio of the number of cases in the first cluster to the total number of cases in the historical case set. The first universality index reflects the representativeness and universality of the first case in the entire case set. For example, assuming that there are 1,000 historical cases in total, the historical case set is divided into 10 clusters through cluster analysis, and the first cluster contains 100 cases. Therefore, the first universality index is 100 / 1,000=0.1, indicating that the case is less popular among all historical cases.

[0093] S25 may specifically be: reading preset discussion features, which may include the number of discussants and the length of discussion on any platform, where any platform may refer to any platform in the entire network platform set. Further feature collection is then performed on the first historical case record based on the preset discussion features to obtain a second feature data set. The feature collection at this stage is intended to assess the discussion popularity of the case in society or academia, thereby providing data support for subsequent educational enlightenment analysis. For example, assuming that on a certain social platform, the number of discussants for the case is 500 and the discussion duration is 200 hours, these data are constructed into a second feature data set.

[0094] S26 may specifically include: performing standardized coefficient of variation weighting processing on the second characteristic data set. Standardized coefficient of variation weighting processing is a key step in calculating the revelation index or educational revelation index. The data in the second characteristic data set is standardized using existing data standardization methods to eliminate differences between different characteristic dimensions. Then, the coefficient of variation is used to weight this data to obtain a comprehensive first revelation index or first educational revelation index. If the discussion of the case is outstanding among historical cases, the coefficient of variation will be weighted and increased, ultimately resulting in a high educational revelation index, reflecting the importance and inspiration of the case in education.

[0095] S27 may specifically be: weighting the first universality index and the first revelatory index to calculate the first typical index, wherein the weights of the first universality index and the first revelatory index may be set by professional and technical personnel in this field based on actual needs, and this specification does not limit this.

[0096] S28 may specifically be: the first typical index may be compared with a preset typical threshold (preset typical index threshold), and when the first typical index reaches the preset typical threshold, the first case may be regarded as a historical typical case. At this point, it is indicated that the first case is outstanding in terms of universality and educational enlightenment, and meets the standards of a typical case, and is included in the historical case knowledge base. Among them, the above-mentioned predetermined screening strategy may be a pre-set case screening standard, such as a preset typical threshold, and the preset typical threshold may be a threshold corresponding to when it is considered a typical case, which may be set according to actual needs, and this specification does not specifically limit this.

[0097] For example, the calculated first case's first universality index is 0.8, the first educational revelation index is 0.7, and the first typicality index is 0.75 (based on average weighted calculation). If the preset typicality threshold is 0.7, the case is considered a typical case and added to the historical case knowledge base. Through the above steps, we can ensure that the cases with the greatest educational value and social impact can be screened from a large number of historical cases for use in intelligent question answering and analysis.

[0098] In some embodiments, the historical case knowledge base in S103 is constructed by screening typical historical cases, and in specific implementation, it may include:

[0099] S29: Using the first feature dataset as a first label for the first case, and obtaining a key knowledge point label corresponding to the first case as a second label;

[0100] S30: Construct a historical case knowledge base based on the first case, the first label, and the second label.

[0101] Specifically, the first case can be assigned a first label. This label comes from a first feature dataset, which is formed by extracting various data from the first case, such as the field, object, and case type involved in the case. The first label can be considered a classification label for the case, representing the overall characteristics of the case and helping the question-answering system identify the type of the case. In addition to the first label, a second label is also extracted. The second label identifies the key knowledge point in the case, that is, the case is associated with a key point. The specific label can be determined based on the educational content involved in the case. Ultimately, a historical case knowledge base can be constructed based on the first case, the first label, and the second label.

[0102] In some embodiments, the real-time case knowledge base in S103 above can be constructed in the following manner:

[0103] S31: collecting features of the second case based on the preset case features to obtain a third feature data set, where the second case is a case at the first time in the real-time key knowledge point case;

[0104] S32: Using the third feature data set as a third label for the second case, and obtaining a key knowledge point label corresponding to the second case as a fourth label;

[0105] S33: Construct a real-time case knowledge base based on the second case, the third label, and the fourth label.

[0106] Specifically, the above-mentioned real-time key knowledge point cases can also be called time series data of real-time key knowledge point cases, which usually come from various social platforms, news media, education discussion forums and other channels. These real-time case records are dynamically obtained through existing crawler technology, API interfaces or data partners, etc., to obtain the time series data of real-time key knowledge point cases. These data reflect the current discussions and social hot issues in real time. The construction of the above-mentioned real-time case knowledge base depends on the time series number of real-time key knowledge point cases. The case at the first time in the time series data of the real-time key knowledge point case can be used as the second case, and the first time can be any time in the time series data.

[0107] Specifically, the second case can be first feature extracted according to predetermined case characteristics to obtain a third feature dataset, specifically including the case domain, object, and type. This third feature dataset can be used as a third label for the second case, representing its specific characteristics. For example, the third label might be a topic label for mental health education. The key knowledge point labels corresponding to the second case can then be used as the fourth label for the second case. Based on the second case, the third label, and the fourth label, a real-time case knowledge base can be constructed. This real-time case knowledge base now stores all labeled real-time case data and can promptly respond to queries and analysis requests for the latest events. Unlike historical case knowledge bases, real-time case knowledge bases are characterized by high data mobility and frequent updates. It is important to ensure that the real-time case knowledge base can efficiently process the constantly updated real-time data and quickly respond to user queries when needed. Through the real-time case knowledge base, the question-answering system can provide timely answers to current social hot topics and related educational content, ensuring that users receive the most timely and relevant information.

[0108] Finally, a dual knowledge base can be formed based on the historical case knowledge base and the real-time case knowledge base, forming a complementary system structure. This not only leverages the experience of historical cases but also enables timely responses to emerging issues and social events. During the dual knowledge base integration process, the two knowledge bases can be connected, allowing the historical case knowledge base and the real-time case knowledge base to call upon and complement each other. Through the collaborative operation of the dual knowledge base, it is possible to achieve two-way access to historical experience and real-time events, providing both classic cases from the past and rapid responses to new issues in the present, forming a dynamically updated, comprehensive intelligent question-answering system.

[0109] In some embodiments, the above-mentioned step S103 of traversing the first key knowledge point in the target key knowledge point set in the pre-built dual knowledge base to obtain a first answer case list may include:

[0110] S34: Traversing the historical case knowledge base in the pre-built dual knowledge base for the first target key knowledge point to obtain a first historical case set;

[0111] S35: Traversing the real-time case knowledge base in the pre-built dual knowledge base for the first target key knowledge point to obtain a first real-time case set;

[0112] S36: combining the first historical case set and the first real-time case set to obtain a first candidate case set for a first target key knowledge point;

[0113] S37: Calculating the matching degree between the first candidate case in the first candidate case set and the first target key knowledge point;

[0114] S38: Sort the first candidate cases according to the matching degree calculation result, and select target candidate cases from the sorting result as the first answer case list, wherein the sorting threshold of the target candidate cases is greater than the preset sorting threshold.

[0115] Specifically, when the first target key knowledge point is matched, the cases related to the first target key knowledge point can be found by traversing the dual knowledge base, and a first answer case list can be generated. The core of this process is to use the historical classic cases and real-time cases in the dual knowledge base, combined with the first target key knowledge point, to screen out the most relevant cases to ensure the pertinence and timeliness of the reply content. Through this dual knowledge base traversal analysis and case screening, users can be provided with detailed, comprehensive and accurate replies. Among them, the process of obtaining the first answer case list is the above-mentioned S34-S38, which is as follows:

[0116] S34 may specifically be: performing a traversal analysis on the first target key knowledge point in the historical case knowledge base, retrieving all cases involving the first target key knowledge point in the historical case knowledge base, and forming a first historical case set.

[0117] S35 may specifically be: performing the same traversal analysis in the real-time case knowledge base, searching for real-time cases related to the first target key knowledge point, and forming a first real-time case set.

[0118] S36 may specifically be: combining the first historical case set and the first real-time case set to form a complete case set as the first candidate case set for the first target key knowledge point, which includes all historical cases and real-time cases related to the first target key knowledge point, and represents the application and discussion of the key knowledge point at different times and in different contexts.

[0119] S37 may specifically include calculating a matching degree between a first candidate case in the first candidate case set and a first target key knowledge point, wherein the first candidate case may generally refer to any case in the first candidate case set. The matching degree calculation is intended to evaluate the relevance between the first candidate case and the first target key knowledge point.

[0120] S38 can be specifically as follows: after calculating the matching degree of the first candidate case, all first candidate cases are sorted in descending order according to the matching degree calculation result to obtain the sorting result. According to the preset rules, a predetermined proportion of cases (i.e., target candidate cases) can be taken from the sorting result as the final first answer case list. Among them, the predetermined proportion is the preset sorting threshold mentioned above, which can be set according to actual needs. For example, the top 10% of the cases in terms of matching degree can be selected, or the top 5 most relevant cases can be selected. The goal of this step is to ensure that the question-answering system only displays the most representative and relevant cases, avoid information overload, and help users quickly find the most relevant answers to their questions.

[0121] Therefore, by traversing historical and real-time cases across the dual knowledge bases, combined with matching calculations and descending sorting, we effectively filter out the most relevant cases for user reference. This precise matching and sorting mechanism provides users with the most valuable and targeted case responses, ensuring the quality and efficiency of answering questions.

[0122] In some embodiments, the specific process of calculating the matching degree between the first candidate case in the first candidate case set and the first target key knowledge point in S37 may be as follows:

[0123] S371: Analyze the first candidate case row, extract all key knowledge point labels involved in the first candidate case, and obtain the first candidate key knowledge point labels;

[0124] S372: Calculate the first candidate coverage between the first candidate key knowledge point label and the target key knowledge point set;

[0125] S373: The first candidate coverage is used as the matching degree between the first candidate case and the first target key knowledge point.

[0126] Among them, the core of calculating the first candidate coverage in S372 is to evaluate the overlap between each candidate key knowledge point label and the target key knowledge point set. Specifically, the first candidate key knowledge point label is matched with the label in the target key knowledge point set, and the ratio of the number of overlapping labels to the total number of key knowledge points in the target key knowledge point set is calculated. The result is the first candidate coverage. The first candidate coverage can be used in S373 as the matching degree between the first candidate case and the first target key knowledge point (which can be called the first matching degree). The first matching degree represents the degree of correlation between the key knowledge point label of the candidate case and the target key knowledge point set. The higher the value, the more the case meets the user's needs.

[0127] By calculating the coverage of the first candidate key knowledge point labels, the relevance of each candidate case can be accurately evaluated, thereby providing users with more accurate and targeted responses, effectively improving the matching accuracy of the intelligent question-answering system, and ensuring that users can get the response content that best meets their needs.

[0128] In some embodiments, the above-mentioned determination of whether the type of the target question answering requirement is the first preset requirement type may further include:

[0129] If not, the extracted target keywords are traversed in the dual knowledge base to obtain a second answer case list;

[0130] The visualized second answer case list is displayed on a target display of a target user.

[0131] Specifically, it can be determined whether the target keyword contains the first professional word identifier. If not, it is inferred that the target question-answering demand does not fully meet the set first preset demand type, and may be the second preset demand type. In order to deal with this situation, the target user's target question-answering demand can be directly analyzed and screened through the historical case knowledge base and the real-time case knowledge base in the dual knowledge base, and the target keywords are traversed and analyzed in the dual knowledge base, that is, the core keywords in the question are matched and analyzed. The cases in the dual knowledge base usually already contain various cases, discussions and social events related to key knowledge points. Therefore, through this keyword matching, a series of case collections related to the target user's target question-answering demand can be found. These cases can be composed of the historical case knowledge base and the real-time case knowledge base to generate a second answer case list. The second answer case list is visualized and displayed to the target user through a target display (such as a user interface, a smart screen, etc.).

[0132] Therefore, by displaying all cases related to user problems in an intuitive and structured way, the user experience can be improved and timely and efficient services can be provided in response to changing educational needs.

[0133] S104: Displaying the visualized first answer case list to the target display of the target user.

[0134] Specifically, the first answer case list selected from the dual knowledge base can be visualized. After the visualization is completed, the first answer case list is displayed on the target display of the target user in the form of graphics, images, tables, charts, etc. The target display can be the user's computer, tablet, smartphone, teaching screen, or other display device.

[0135] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. For details, please refer to the description of the aforementioned related processing embodiments, and no further description is given here.

[0136] The above describes the present invention. However, it is worth noting that this specific embodiment is only intended to better illustrate the present application and to describe specific embodiments of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0137] The above method is described below in conjunction with a specific embodiment. However, it should be noted that this specific embodiment is only for better illustrating the present application and does not constitute an improper limitation to the present application.

[0138] Before implementation, a pre-defined screening strategy was introduced to analyze and screen historical key knowledge point case records to obtain historical typical cases and build a historical case knowledge base. A real-time case knowledge base was constructed through the time sequence of dynamically acquired real-time key knowledge point cases, and this knowledge base was combined with the historical case knowledge base to form a dual knowledge base.

[0139] In a specific implementation, a target user's target Q&A requirement is obtained, and it is determined whether the target Q&A requirement is of a first preset requirement type. If so, a target key knowledge point set corresponding to the target Q&A requirement is matched, where the target key knowledge point set includes a first target key knowledge point. The first target key knowledge point is traversed and analyzed within the dual knowledge base to obtain a first list of answer cases. The visualized first list of answer cases is displayed on the target display of the target user.

[0140] By introducing dynamically acquired real-time case data on key knowledge points and combining it with historical case studies, we build a dynamic and intelligent dual knowledge base that efficiently integrates the historical knowledge base and real-time data streams. Furthermore, by matching target key knowledge points with professional terms, we intelligently filter the most relevant case studies based on the user's background and needs. This effectively integrates education into professional course teaching, achieving an organic combination of subject knowledge and education, thereby achieving the technical effect of enhancing the depth of responses.

[0141] Although this specification provides examples such as the following examples or the accompanying Figure 2 The method operation steps or device structure shown, but based on routine or no creative labor, the method or device may include more or fewer operation steps or module units after partial merger. In the steps or structures where there is no necessary causal relationship logically, the execution order of these steps or the module structure of the device is not limited to the execution order or module structure shown in the embodiments or drawings of this specification. When the method or module structure described is applied to an actual device, server or terminal product, it can be executed sequentially or in parallel according to the method or module structure shown in the embodiments or drawings (for example, a parallel processor or multi-threaded processing environment, or even a distributed processing, server cluster implementation environment). Based on the above-mentioned intelligent question-answering method based on the collaboration of dual knowledge bases, the embodiments of this specification also propose an embodiment of an intelligent question-answering device based on the collaboration of dual knowledge bases. As Figure 2 As shown, the device may specifically include the following modules:

[0142] The judgment module 201 may be used to obtain a target question-answering requirement of a target user and determine whether the type of the target question-answering requirement is a first preset requirement type;

[0143] The matching module 202 may be configured to match a target key knowledge point set corresponding to the target question answering requirement.

[0144] The traversal module 203 may be configured to traverse a first target key knowledge point in the target key knowledge point set in a pre-built dual knowledge base to obtain a first answer case list, wherein the dual knowledge base is composed of a historical case knowledge base and a real-time case knowledge base, wherein the historical case knowledge base is constructed from screened historical typical cases;

[0145] The display module 204 may be configured to display the visualized first answer case list to a target display of a target user.

[0146] In some embodiments, the above-mentioned judgment module 201 can be specifically used to extract the target keywords in the target question text corresponding to the target question and answer demand; determine whether the target keywords match the first professional word identifier corresponding to the first course, and if so, determine that the type of the target question and answer demand is the first preset demand type, and the first course is any course in the pre-acquired target user course list. The type of the target question and answer demand is the first preset demand type, indicating that the target question and answer demand is related to the first course.

[0147] In some embodiments, the matching module 202 may be specifically configured to match a target key knowledge point set corresponding to a first course related to the target question-answering requirement.

[0148] In some embodiments, the above-mentioned traversal module 203 can be specifically used to extract the first historical case record corresponding to the first case in the historical key knowledge point case record; read the preset case feature, and collect features of the first historical case record based on the preset case feature to obtain a first feature data set; cluster the historical case set in the historical key knowledge point record with the first feature data set as the clustering constraint to obtain a first cluster cluster, and the historical case set is composed of multiple first cases; determine the first universality index of the first case according to the ratio of the number of cases in the first cluster cluster to the total number of cases in the historical case set; read the preset discussion feature, and collect features of the first historical case record based on the preset discussion feature to obtain a second feature data set; perform standardized coefficient of variation weighting processing on the second feature data set to obtain a first revelation index of the first case; perform weighted processing on the first universality index and the first revelation index to obtain a first typical index of the first case; when the first typical index reaches a preset typical threshold, the first case is used as a historical typical case.

[0149] In some embodiments, the traversal module 203 can also be used to use the first feature data set as the first label of the first case, and obtain the key knowledge point label corresponding to the first case as the second label; and construct a historical case knowledge base based on the first case, the first label, and the second label.

[0150] In some embodiments, the above-mentioned traversal module 203 can also be used to collect features of the second case based on preset case features to obtain a third feature data set, where the second case is a case at the first time in the real-time key knowledge point case; the third feature data set is used as the third label of the second case, and the key knowledge point label corresponding to the second case is obtained as the fourth label; a real-time case knowledge base is constructed based on the second case, the third label, and the fourth label.

[0151] In some embodiments, the above-mentioned traversal module 203 can also be specifically used to traverse the historical case knowledge base in the pre-built dual knowledge base for the first target key knowledge point to obtain a first historical case set; traverse the real-time case knowledge base in the pre-built dual knowledge base for the first target key knowledge point to obtain a first real-time case set; combine the first historical case set with the first real-time case set to obtain a first candidate case set for the first target key knowledge point; calculate the matching degree between the first candidate case in the first candidate case set and the first target key knowledge point; sort the first candidate case according to the matching degree calculation result, and filter the target candidate case from the sorting result as the first answer case list, and the sorting threshold of the target candidate case is greater than the preset sorting threshold.

[0152] In some embodiments, the judgment module 201 can be further used to traverse the extracted target keywords in the dual knowledge base to obtain a second answer case list if no; and display the visualized second answer case list to the target display of the target user.

[0153] As can be seen from the above, the intelligent question-answering device based on dual knowledge base collaboration provided in the embodiments of this specification can effectively improve the depth of answers and enable target users to obtain their desired answers.

[0154] An embodiment of this specification also provides an electronic device based on the above-mentioned intelligent question-answering method based on the collaboration of dual knowledge bases, including a processor and a memory for storing processor executable programs / instructions. When the processor is specifically implemented, it can perform the following steps according to the program / instructions: obtain the target question-answering needs of the target user, and determine whether the type of the target question-answering needs is a first preset need type; if so, match the target key knowledge point set corresponding to the target question-answering needs; traverse the first target key knowledge point in the target key knowledge point set in a pre-built dual knowledge base to obtain a first answer case list, the dual knowledge base consists of a historical case knowledge base and a real-time case knowledge base, and the historical case knowledge base is constructed by screened historical typical cases; and display the visualized first answer case list to the target display of the target user.

[0155] In order to complete the above instructions more accurately, refer to Figure 3 As shown, the embodiment of this specification also provides another specific electronic device, wherein the electronic device includes a network communication port 301, a processor 302 and a memory 303, and the above structures are connected through internal cables so that each structure can perform specific data interaction.

[0156] The network communication port 301 may be used to obtain a target question-answering requirement of a target user and determine whether the target question-answering requirement is a first preset requirement type.

[0157] The processor 302 may be specifically configured to: match a target key knowledge point set corresponding to the target question answering requirement; traverse a first target key knowledge point in the target key knowledge point set in a pre-built dual knowledge base to obtain a first answer case list, wherein the dual knowledge base is composed of a historical case knowledge base and a real-time case knowledge base, wherein the historical case knowledge base is constructed from screened historical typical cases; and display the visualized first answer case list on a target display of a target user;

[0158] The memory 303 may be specifically used to store corresponding instruction programs.

[0159] In this embodiment, the network communication port 301 can be a virtual port that is bound to different communication protocols, thereby being capable of sending or receiving different data. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. Furthermore, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM or CDMA; it can also be a Wi-Fi chip; or it can be a Bluetooth chip.

[0160] In this embodiment, the processor 302 can be implemented in any suitable manner. For example, the processor can take the form of a microprocessor or a processor and a computer-readable medium storing computer-readable program code (such as software or firmware) executable by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, an embedded microcontroller, etc. This specification is not limited to this.

[0161] In this embodiment, the memory 303 may include multiple levels. In a digital system, anything that can store binary data can be a memory. In an integrated circuit, a circuit with a storage function that does not have a physical form is also called a memory, such as RAM, FIFO, etc. In a system, a storage device with a physical form is also called a memory, such as a memory stick, TF card, etc.

[0162] An embodiment of this specification also provides a computer storage medium based on the above-mentioned intelligent question-answering method based on the collaboration of dual knowledge bases, wherein the computer storage medium stores a computer program / instruction, which, when executed, implements: obtaining the target question-answering demand of the target user, and determining whether the type of the target question-answering demand is a first preset demand type; if so, matching a target key knowledge point set corresponding to the target question-answering demand; traversing the first target key knowledge point in the target key knowledge point set in a pre-constructed dual knowledge base to obtain a first answer case list, wherein the dual knowledge base is composed of a historical case knowledge base and a real-time case knowledge base, and the historical case knowledge base is constructed by screened historical typical cases; and displaying the visualized first answer case list to the target display of the target user.

[0163] In this embodiment, the storage medium includes, but is not limited to, random access memory (RAM), read-only memory (ROM), cache, hard disk drive (HDD), or memory card. The memory can be used to store computer program instructions. The network communication unit can be an interface configured in accordance with the standards specified by the communication protocol for network connection communication.

[0164] In this embodiment, the functions and effects specifically implemented by the program instructions stored in the computer storage medium can be explained in comparison with other implementations and will not be repeated here.

[0165] Although this specification provides the method operation steps as described in the embodiments or flow charts, more or fewer operation steps may be included based on conventional or non-creative means. The order of steps listed in the embodiments is only one way of executing the order of many steps and does not represent the only execution order. When the device or client product in practice is executed, it can be executed in sequence or in parallel according to the method shown in the embodiments or the drawings (for example, a parallel processor or a multi-threaded processing environment, or even a distributed data processing environment). The term "comprise", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, product or device including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such process, method, product or device. In the absence of more restrictions, it is not excluded that there are other identical or equivalent elements in the process, method, product or device including the elements. Words such as first and second are used to represent names and do not represent any particular order.

[0166] Those skilled in the art will also appreciate that, in addition to implementing the controller in pure computer-readable program code, it is entirely possible to implement the same functionality by logically programming the method steps in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, embedded microcontrollers, and the like. Therefore, such a controller can be considered a hardware component, and the devices included therein for implementing various functions can also be considered structures within the hardware component. Alternatively, the devices for implementing various functions can be considered both software modules implementing the method and structures within the hardware component.

[0167] This specification may be described in the general context of computer-executable instructions, such as program modules, executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, classes, and the like that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media, including storage devices.

[0168] Through the description of the above embodiments, it can be seen that those skilled in the art can clearly understand that this specification can be implemented by means of software plus the necessary general hardware platform. Based on this understanding, the technical solution of this specification can essentially be embodied in the form of a software product. This computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a mobile terminal, a server, or a network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of this specification.

[0169] The various embodiments in this specification are described in a progressive manner. References to the common or similar parts of the various embodiments are sufficient. Each embodiment focuses on the differences from the other embodiments. This specification can be used in a variety of general-purpose or specialized computer system environments or configurations. For example, personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments that include any of the above systems or devices.

[0170] Although the present specification has been described with reference to the embodiments, persons skilled in the art will appreciate that there are many variations to the present specification without departing from the spirit of the present specification, and it is intended that the appended claims encompass such variations without departing from the spirit of the present specification.

Claims

1. An intelligent question answering method based on dual knowledge base collaboration, characterized in that: include: Obtaining a target question-answering requirement of a target user, and determining whether a type of the target question-answering requirement is a first preset requirement type; If yes, match the target key knowledge point set corresponding to the target question answering requirement; Traversing the first target key knowledge point in the target key knowledge point set in a pre-built dual knowledge base to obtain a first answer case list, wherein the dual knowledge base is composed of a historical case knowledge base and a real-time case knowledge base, and the historical case knowledge base is constructed by screening historical typical cases; Displaying the visualized first answer case list to a target display of a target user; The typical historical cases are selected in the following way: Extracting the first historical case record corresponding to the first case in the historical key knowledge point case record; Reading preset case features, and collecting features of the first historical case record based on the preset case features to obtain a first feature data set; Using the first feature data set as a clustering constraint, clustering a set of historical cases in the historical key knowledge point records to obtain a first cluster, where the set of historical cases consists of a plurality of first cases; determining a first universality index of the first case according to a ratio of the number of cases in the first cluster to the total number of cases in the historical case set; Reading a preset discussion feature, and collecting features of the first historical case record based on the preset discussion feature to obtain a second feature data set; performing standardized coefficient of variation weighting processing on the second feature data set to obtain a first revealing index of the first case; performing weighted processing on the first universality index and the first revelatory index to obtain a first typicality index of the first case; When the first typical index reaches a preset typical threshold, the first case is regarded as a historical typical case.

2. The method according to claim 1, characterized in that The determining whether the type of the target question answering requirement is a first preset requirement type includes: Extract target keywords from the target question text corresponding to the target question answering requirement; Determine whether the target keyword matches a first professional word identifier corresponding to the first course; if so, determine that the type of the target question-answering demand is a first preset demand type, the first course is any course in a pre-acquired target user course list, and the type of the target question-answering demand being the first preset demand type indicates that the target question-answering demand is related to the first course; Accordingly, the target key knowledge point set corresponding to the target question answering requirement is matched, including: Match the target key knowledge point set corresponding to the first course related to the target question answering requirement.

3. The method according to claim 1, characterized in that The historical case knowledge base is constructed from screened historical typical cases, including: Using the first feature data set as a first label for the first case, and obtaining a key knowledge point label corresponding to the first case as a second label; A historical case knowledge base is constructed based on the first case, the first label, and the second label.

4. The method according to claim 1, wherein The real-time case knowledge base is constructed in the following manner: Based on the preset case features, features of the second case are collected to obtain a third feature data set, where the second case is a case at the first time in the real-time key knowledge point case; Using the third feature data set as a third label for the second case, and obtaining a key knowledge point label corresponding to the second case as a fourth label; A real-time case knowledge base is constructed based on the second case, the third label, and the fourth label.

5. The method according to claim 1, wherein The first target key knowledge point in the target key knowledge point set is traversed in the pre-built dual knowledge base to obtain a first answer case list, including: Traversing the historical case knowledge base in the pre-built dual knowledge base for the first target key knowledge point to obtain a first historical case set; Traversing the real-time case knowledge base in the pre-built dual knowledge base for the first target key knowledge point to obtain a first real-time case set; Combining the first historical case set with the first real-time case set to obtain a first candidate case set for a first target key knowledge point; Calculate the matching degree between the first candidate case in the first candidate case set and the first target key knowledge point; According to the matching degree calculation result, the first candidate cases are sorted, and target candidate cases are screened from the sorting results as the first answer case list, where the sorting threshold of the target candidate cases is greater than the preset sorting threshold.

6. The method according to claim 1, characterized in that The method further comprises: If not, the extracted target keywords are traversed in the dual knowledge base to obtain a second answer case list; The visualized second answer case list is displayed on a target display of a target user.

7. An intelligent question-answering device based on dual knowledge base collaboration, characterized in that: include: A judgment module, configured to obtain a target question-answering requirement of a target user and judge whether the type of the target question-answering requirement is a first preset requirement type; A matching module, configured to match a target key knowledge point set corresponding to the target question answering requirement; a traversal module, configured to traverse a first target key knowledge point in the target key knowledge point set in a pre-built dual knowledge base to obtain a first answer case list, wherein the dual knowledge base is composed of a historical case knowledge base and a real-time case knowledge base, and the historical case knowledge base is constructed from screened historical typical cases; A display module, configured to display the visualized first answer case list to a target display of a target user; The typical historical cases are selected in the following way: Extracting the first historical case record corresponding to the first case in the historical key knowledge point case record; Reading preset case features, and collecting features of the first historical case record based on the preset case features to obtain a first feature data set; Using the first feature data set as a clustering constraint, clustering a set of historical cases in the historical key knowledge point records to obtain a first cluster, where the set of historical cases consists of a plurality of first cases; determining a first universality index of the first case according to a ratio of the number of cases in the first cluster to the total number of cases in the historical case set; Reading a preset discussion feature, and collecting features of the first historical case record based on the preset discussion feature to obtain a second feature data set; performing standardized coefficient of variation weighting processing on the second feature data set to obtain a first revealing index of the first case; performing weighted processing on the first universality index and the first revelatory index to obtain a first typicality index of the first case; When the first typical index reaches a preset typical threshold, the first case is regarded as a historical typical case.

8. An electronic device, characterized in that: include: A memory and a processor, wherein the processor and the memory are communicatively connected to each other, the memory stores computer instructions, and the processor implements the steps of the method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images