CN110457455B - Ternary logic question-answer consultation optimization method, system, medium and equipment - Google Patents
Ternary logic question-answer consultation optimization method, system, medium and equipment Download PDFInfo
- Publication number
- CN110457455B CN110457455B CN201910674219.2A CN201910674219A CN110457455B CN 110457455 B CN110457455 B CN 110457455B CN 201910674219 A CN201910674219 A CN 201910674219A CN 110457455 B CN110457455 B CN 110457455B
- Authority
- CN
- China
- Prior art keywords
- logic
- nodes
- node
- semantic
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3346—Query execution using probabilistic model
Abstract
The invention provides a three-valued logic question-answer consultation optimization method, a system, a medium and equipment, which respectively perform semantic analysis and logic analysis aiming at a pre-acquired knowledge base, and respectively acquire semantic nodes and logic nodes; calculating three-value logic operation values of the logic nodes and the semantic nodes according to a pre-acquired query text, and counting the jump probability of the semantic nodes; calculating the probability of each node for obtaining an exact answer according to the jump probability and the three-value logic operation value; obtaining an optimal consultation question-answer node according to the probability of obtaining the exact answer by each node; the invention can effectively improve the accuracy and efficiency of natural language processing under the logic context.
Description
Technical Field
The invention relates to the field of natural language processing, in particular to a three-value logic question-answer consultation optimization method, system, medium and equipment.
Background
In traditional semantic concepts that include logic, true or false is often used to define or interpret a preset, referred to as a semantic preset or a preset semantic interpretation. In practice, however, the interpretation of the semantics is not just false or true. Although the application of ternary logic to describe uncertain components in semantics has been proposed in the early days, how to apply ternary logic to natural language processing to improve the accuracy and efficiency of natural language recognition in the context of logic remains to be studied.
Disclosure of Invention
In view of the problems in the prior art, the invention provides a ternary logic question-answering consulting optimization method, system, medium and equipment, and mainly solves the problem that logic description improves natural language processing efficiency and precision.
In order to achieve the above and other objects, the present invention adopts the following technical solutions.
A three-value logic question-answer consultation optimization method comprises the following steps:
respectively performing semantic analysis and logic analysis on a pre-acquired knowledge base to respectively acquire semantic nodes and logic nodes;
calculating three-value logic operation values of the logic nodes and the semantic nodes according to a pre-acquired query text, and counting the jump probability of the semantic nodes;
calculating the probability of each node for obtaining an exact answer according to the jump probability and the three-value logic operation value;
and obtaining the optimal consultation question-answer node according to the probability of obtaining the exact answer by each node.
Optionally, the semantic analysis comprises:
extracting semantic features of texts in the knowledge base;
and constructing semantic nodes according to the semantic features, and establishing semantic connection among the semantic nodes according to the context relationship.
Optionally, the logical analysis comprises:
extracting the logic characteristics of the text in the knowledge base;
constructing a logic node according to the logic characteristics;
establishing logical connection between the logical nodes according to the logical relation among the logical characteristics;
and establishing the connection relation between the logic node and the semantic node according to context semantics.
Optionally, the calculating the three-valued logic operation values of the logic node and the semantic node according to the pre-obtained query text specifically includes:
extracting key features of the query text;
comparing the key features with the semantic nodes and the logic nodes respectively, and judging logic values corresponding to the semantic nodes and the logic nodes respectively according to a preset three-value logic rule;
and calculating to obtain a logic operation value between the logic nodes according to the logic true value of the logic nodes and by combining the connection relation between the logic nodes.
Optionally, the preset three-value logic rule specifically includes:
classifying the semantic nodes and the logic nodes together, wherein the classification types comprise semantic type nodes, numerical type nodes, Boolean type nodes and character string type nodes;
when the key features have features associated with the semantic type nodes, if the similarity between the semantics of the semantic type nodes and the corresponding features in the key features is higher than a set similarity threshold, the logic value of the semantic type nodes is true, otherwise, the logic value of the semantic type nodes is false, and when the features associated with the semantic type nodes do not exist in the key features, the logic value of the semantic type nodes is unknown;
when the semantic type node connected with the numerical type node is true, if a numerical value corresponding to the numerical type node exists in the key features, the logical value of the numerical type node is true, otherwise, the logical value of the numerical type node is unknown;
when the semantic node connected with the Boolean type node is true, if verification features corresponding to the Boolean type node exist in the key features, the logic value of the Boolean type node is true, otherwise, the logic value of the Boolean type node is false, and when verification contents corresponding to the Boolean type node do not exist in the key features, the logic value of the Boolean type node is unknown;
when the semantic type node connected with the character string type node is true, if the character string corresponding to the character string type node exists in the key feature, the logical value of the character string type node is true, otherwise, the logical value of the character string type node is unknown.
Optionally, a formula for calculating that the three-valued logic values corresponding to the semantic node and the logic node are unknown is as follows:
wherein R (k) represents the logical value of node k, LPr (i) represents the set of all said logical nodes connected to node i, Pr (i) represents the set of all semantic nodes connected to node i, LkiRepresenting the logical operation values of node k and node i, SkiRepresenting the probability of a hop between nodes k and i.
Optionally, the calculating a probability of each node obtaining an exact answer is specifically represented as:
wherein, c1And c2Is a weight, c1+c2=1，LC(i) Representing a set of all next-level said logical nodes connected to node i; c (i) represents a set of all next-level nodes connected to node i; l isijRepresenting the logical operation values of nodes i to j; sijIndicating the hop probability of node i jumping to node j.
A three-valued logic question-answering consulting optimization system, comprising:
the node creation module is used for respectively performing semantic analysis and logic analysis on a pre-acquired knowledge base to respectively acquire semantic nodes and logic nodes;
the logic calculation module is used for calculating the three-value logic calculation values of the logic nodes and the semantic nodes according to the pre-acquired inquiry text;
the probability calculation module is used for counting the skipping probability of the semantic nodes; calculating the probability of each node for obtaining an exact answer according to the jump probability and the three-value logic operation value;
and the optimizing module is used for acquiring the optimal consultation question-answering node according to the probability of acquiring the exact answer by each node.
A computer-readable storage medium, in which a computer program is stored, which, when loaded and executed by a processor, implements the three-valued logic question-answer consulting optimization method.
An apparatus comprising a processor and a memory; wherein the content of the first and second substances,
the memory is used for storing a computer program;
the processor is used for loading and executing the computer program, so that the equipment executes the three-value logic question-answer consultation optimization method.
As described above, the present invention provides a method, system, medium, and apparatus for optimizing a three-valued logic question-answering consultation, which has the following advantages.
The logic node and the semantic node are combined by calculating the logic probability and the jump probability, so that the accuracy of natural language processing is improved; the addition of logic operation can make the optimizing result more quickly converge and raise optimizing efficiency.
Drawings
Fig. 1 is a flowchart of a three-valued logic question-answer consultation optimization method in an embodiment of the invention.
FIG. 2 is a block diagram of a three-valued logic query-answering optimization system in accordance with an embodiment of the present invention.
Fig. 3 is a schematic structural diagram of an apparatus according to an embodiment of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
Referring to fig. 1, the present disclosure provides a three-valued logic query and answer optimization method, which includes steps S01-S04.
In step S01, semantic analysis and logical analysis are performed on the knowledge base acquired in advance, and a semantic node and a logical node are acquired:
in an embodiment, the knowledge base can collect relevant technical documents of corresponding fields according to different technical fields, and the relevant technical documents are sorted to obtain the knowledge base. Taking the legal field as an example, the information of legal regulations, legal forums, related legal treatises, magazines and the like is collected and arranged and is input into a computer database to form a knowledge base.
When semantic analysis and logic analysis are carried out, word segmentation, sentence segmentation and segmentation processing can be carried out on the text in the knowledge base, and the text is divided into relatively independent words, sentences and paragraphs. Semantic features and logical features are then extracted from the processed text. The semantic features herein mainly refer to words, numerical values, or character strings containing specific semantics, and the logical features mainly refer to logical constraint relationships in the semantic features, including logical relationships, conditional judgment relationships, arithmetic operation relationships, whether they are true or false, and the like.
In one embodiment, the semantic features are converted into word vectors and input into a recognition model to train a question-answering network, the semantic features are used as semantic nodes of the question-answering network, and the semantic connections among the semantic nodes are constructed by combining the context semantic relations of texts corresponding to the semantic features. And meanwhile, the logic characteristics are used as logic nodes of the question-answering network, the logic nodes are in logic connection with the corresponding semantic nodes, and the logic connection between the logic nodes is established by combining the logic relation between the logic characteristics. And constructing a complete question-answering network through the logic nodes and the semantic nodes.
In step S02, the three-valued logic operation values of the logic node and the semantic node are calculated based on the query text acquired in advance, and the hop probability of the semantic node is counted:
in one embodiment, the query text refers to a query text that is provided by a user for a specific question, and may be a voice text or a directly input text. After the query text is subjected to word segmentation, sentence segmentation and segmentation processing, key features of the query text are extracted, wherein the key features comprise semantics, semantic context relations, logical relations and the like. And comparing the semantic information corresponding to the key features with the semantic nodes of the question-answering network in the step S01, and calculating the similarity between the word vectors corresponding to the semantic information and the word vectors corresponding to the semantic nodes in the question-answering network. The similarity of the word vectors can be obtained by calculating the canonical distance of the word vectors. And when the similarity reaches a set threshold, judging that the semantic information corresponding to the key features is similar to the semantic information corresponding to the semantic nodes in the question-answering network, and taking the similar semantic nodes obtained by calculation as access nodes of the question-answering network. Similarly, the similarity between the corresponding logic information in the key features and the logic nodes is calculated. And taking the access point as a base point, and skipping semantic nodes according to the similarity between the key features of the query text and the nodes corresponding to the question-answering network. When the similarity of the semantic node or the logic node connected with the base point and the corresponding characteristic of the query text reaches a set threshold, the base point jumps to the corresponding semantic node or the logic node, the number of jumping times of the base point is counted, and the ratio of the number of jumping times of the base point to a certain node connected with the base point to the total number of jumping times of the base point is taken as the probability of jumping of the node relative to the base point. For example, if the node a hops to the node B once and the number of hops from the node a to other nodes connected to the node a is N, the hop probability of the node B is 1/N. By the method, the hop probability of other nodes except the base point can be calculated.
In order to facilitate quantitative calculation of nodes in the question-answering network, all the nodes in the question-answering network are divided into four types, namely semantic type nodes, numerical type nodes, Boolean type nodes and character string type nodes. And defines a three-valued logic for quantizing the logical value of each node.
In one embodiment, three-valued logic may be defined as true T, false F, and unknown U, respectively. True is represented by 1, false by 0, and unknown by a value between 0 and 1.
The AND operation that defines the three-valued logic can be expressed as:
where a and B represent the likelihood that different nodes are true, respectively.
TABLE 1
| A B | T | F | U |
| T | T | F | U |
| F | F | F | F |
| U | U | F | U |
Table 1 is a truth table for the logical AND operation between node A and node B.
The OR operation that defines the three-valued logic can be expressed as:
TABLE 2
| A B | T | F | U |
| T | T | T | T |
| F | T | F | U |
| U | T | U | U |
Table 2 is a truth table for the logical OR operation between node A and node B.
The negation operation defining the three-valued logic can be expressed as:
TABLE 3
| A | ^A |
| T | F |
| F | T |
| U | U |
Table 3 is a truth table for the logical NOT operation performed by node A.
Based on the logic operation, according to the node type of the question-answering network, an operation rule of the ternary logic is defined, and the operation rule comprises the following steps:
aiming at the semantic type node, when the key features of the query text have features associated with the semantic type node, if the similarity between the semantics of the semantic type node and the corresponding features in the key features is higher than a set similarity threshold, the logic value of the semantic type node is true, otherwise, the logic value of the semantic type node is false, and when the key features do not have the features associated with the semantic type node, the logic value of the semantic type node is unknown;
for the numerical value type node, when the semantic type node connected with the numerical value type node is true, if a numerical value corresponding to the numerical value type node exists in the key features, the logical value of the numerical value type node is true, otherwise, the logical value of the numerical value type node is unknown;
for a Boolean type node, when a semantic node connected with the Boolean type node is true, if a verification feature corresponding to the Boolean type node exists in the key feature, the logic value of the Boolean type node is true, otherwise the logic value of the Boolean type node is false, and when the verification content corresponding to the Boolean type node does not exist in the key feature, the logic value of the Boolean type node is unknown;
and aiming at the character string type node, when the semantic type node connected with the character string type node is true, if a character string corresponding to the character string type node exists in the key feature, the logical value of the character string type node is true, otherwise, the logical value of the character string type node is unknown.
In an embodiment, the comparison operation under the three-valued logic may be further defined for the value type, including greater than, less than, greater than or equal to, less than or equal to, and equal to. And judging the final result of the logic operation through the constructed comparison function. The comparison function can be set according to the actual context or application scene, when the logic values of the two numerical type nodes are true at the same time, the calculation result of the comparison function is used as the logic operation value, and in other cases, the logic value of one node can be used as the comparison operation value.
The arithmetic operation of three-valued logic can also be defined, including addition, subtraction, multiplication, division and the like, when the logic values of two numerical type nodes are true at the same time, the result of the arithmetic operation is true, otherwise, the logic value of one of the nodes can be used as the value of the arithmetic operation.
Through the defined logic operation mode and logic operation rules, the logic operation value among the nodes of the question-answering network can be calculated.
In an embodiment, the jump probability of the semantic node is counted, the jump frequency can be counted by reading the historical jump records of the semantic node and the child nodes thereof, the jump probability of the semantic node is constructed by a Bayesian probability calculation method, a certain node is taken as a father node, the node connected with the node is taken as a child node, the jump frequency from the father node to the child node is counted, and the jump probability is constructed.
In step S03, the probability of each node obtaining an exact answer is calculated based on the hop probability and the three-valued logic operation value:
according to the logic operation manner and rules in step S02, the possibility that all the logic and arithmetic expressions for constructing the question-answering network are true can be calculated. In calculating the logic operation value, since the logic value unknown U is a number between 0 and 1 for representing uncertainty of the true value of the node, the value of the node whose logic value is unknown can be constructed by the determined true value of the node. The calculation formula can be expressed as:
wherein R (k) represents the logical value of node k, LPr (i) represents the set of all said logical nodes connected to node i, Pr (i) represents the set of all semantic nodes connected to node i, LkiRepresenting the logical operation values of node k and node i, SkiRepresenting the probability of a hop between nodes k and i.
In an embodiment, according to the calculated logical operation value, the probability that each node in the question-answering network can finally obtain an exact answer, that is, P (x) can be further calculatedi). Can be expressed as:
wherein, c1And c2Is a weight, c1+c2=1, lc (i) represents a set of all next-level logical nodes connected to node i; c (i) represents a set of all next-level nodes connected to node i; l isijRepresenting the logical operation values of nodes i to j; sijIndicating the hop probability of node i jumping to node j.
In step S04, an optimal consultation question-answering node is obtained according to the probability that each node obtains an exact answer:
according to the nodes with unknown logical values obtained by calculation in step S03, a set with the unknown nodes as boundaries can be obtained, where all the nodes in the set are semantic type nodes with true logical values, and the boundaries of the semantic type nodes are connected with the nodes with unknown logical values.
And according to the probability values of the nodes obtained by calculation in the step S03, selecting the node with the highest probability value from the set as the optimal consultation question-answering node. The calculation process can be expressed as:
wherein KE represents a set of semantic type nodes whose logical value is true.
According to an embodiment of the present invention, there is also provided a computer storage medium, in which a computer program is stored, and when the computer program is executed, the three-valued logic question-answer consultation optimization method can be implemented. Computer storage media may include any available media for computer storage or data storage devices including one or more available media integrated servers, data centers, and the like. Usable media include magnetic media (e.g., floppy disks, hard disks, magnetic tape), optical media (e.g., DVDs), semiconductor media (e.g., solid state disks), and the like.
Referring to fig. 2, the present embodiment provides a three-valued logic query and answer optimization system for implementing the three-valued logic query and answer optimization method described in the foregoing method embodiments. Since the technical principle of the system embodiment is similar to that of the method embodiment, repeated description of the same technical details is omitted.
In one embodiment, the three-valued logic question-answering consulting optimization system comprises a node creating module 10, a logic calculating module 11, a probability calculating module 12 and an optimizing module 13; the node creation module 10 is configured to assist in executing the step S01 described in the foregoing method embodiment, the logic calculation module 11 is configured to execute the step S02 described in the foregoing method embodiment, the probability calculation module 12 is configured to execute the step S03 in the foregoing method embodiment, and the optimization module 13 is configured to execute the step S04 in the foregoing method embodiment.
Referring to fig. 3, the present embodiment provides an apparatus, which may be a desktop computer, a portable computer, etc., and specifically, the apparatus at least includes a processor 20 and a memory 21.
The processor 20 is configured to perform all or part of the steps of the foregoing method embodiments. The Processor 20 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
In summary, according to the method, the system, the medium and the equipment for consulting and optimizing the trivalued logic question answering, semantic expression can be expanded by describing semantic information and logic relations of nodes through the trivalued logic; the optimization range is determined through the unknown nodes, the convergence of the natural language processing algorithm can be improved, the efficiency is improved, the logic reasoning and the semantic analysis are combined, and the integrity of knowledge can be guaranteed. Therefore, the invention effectively overcomes various defects in the prior art and has high industrial utilization value.
The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.
Claims (9)
1. A three-value logic question-answer consultation optimization method is characterized by comprising the following steps:
respectively performing semantic analysis and logic analysis on a pre-acquired knowledge base to respectively acquire semantic nodes and logic nodes;
calculating the three-value logic operation values of the logic nodes and the semantic nodes according to the pre-acquired query text, and counting the jump probability of the semantic nodes, wherein the method comprises the following steps: extracting key features of the query text; comparing the key features with the semantic nodes and the logic nodes respectively, and judging ternary logic values corresponding to the semantic nodes and the logic nodes respectively according to a preset ternary logic rule; according to the three-valued logic values of the logic nodes, calculating to obtain three-valued logic operation values among the logic nodes by combining the connection relation among the logic nodes; wherein the three-valued logic value comprises: true, false, and unknown;
calculating the probability of each node for obtaining an exact answer according to the jump probability and the three-value logic operation value;
and obtaining the optimal consultation question-answer node according to the probability of obtaining the exact answer by each node.
2. The three-valued logic question-answering consultation optimization method according to claim 1, wherein the semantic analysis comprises:
extracting semantic features of texts in the knowledge base;
and constructing semantic nodes according to the semantic features, and establishing semantic connection among the semantic nodes according to the context relationship.
3. The three-valued logic question-answer consultation optimization method according to claim 1, wherein said logic analysis comprises:
extracting the logic characteristics of the text in the knowledge base;
constructing a logic node according to the logic characteristics;
establishing logical connection between the logical nodes according to the logical relation among the logical characteristics;
and establishing the connection relation between the logic node and the semantic node according to context semantics.
4. The method for consulting and optimizing a trivalued logic question answering according to claim 1, wherein the preset trivalued logic rule specifically includes:
classifying the semantic nodes and the logic nodes together, wherein the classification types comprise semantic type nodes, numerical type nodes, Boolean type nodes and character string type nodes;
when the key features have features associated with the semantic type nodes, if the similarity between the semantics of the semantic type nodes and the corresponding features in the key features is higher than a set similarity threshold, the logic value of the semantic type nodes is true, otherwise, the logic value of the semantic type nodes is false, and when the features associated with the semantic type nodes do not exist in the key features, the logic value of the semantic type nodes is unknown;
when the semantic type node connected with the numerical type node is true, if a numerical value corresponding to the numerical type node exists in the key features, the logical value of the numerical type node is true, otherwise, the logical value of the numerical type node is unknown;
when the semantic node connected with the Boolean type node is true, if verification features corresponding to the Boolean type node exist in the key features, the logic value of the Boolean type node is true, otherwise, the logic value of the Boolean type node is false, and when verification contents corresponding to the Boolean type node do not exist in the key features, the logic value of the Boolean type node is unknown;
when the semantic type node connected with the character string type node is true, if the character string corresponding to the character string type node exists in the key feature, the logical value of the character string type node is true, otherwise, the logical value of the character string type node is unknown.
5. The three-valued logic question-answer consultation optimization method according to claim 4, wherein a formula for calculating the three-valued logic values corresponding to the semantic nodes and the logic nodes as unknown is as follows:
wherein R (k) represents the logical value of node k, LPr (i) represents the set of all said logical nodes connected to node i, Pr (i) represents the set of all semantic nodes connected to node i, LkiRepresenting the logical operation values of node k and node i, SkiRepresenting the probability of a hop between nodes k and i.
6. The method for consulting and optimizing a three-valued logic question-answer according to claim 1, wherein the calculating of the probability of each node obtaining an exact answer is specifically represented as:
wherein, c1And c2Is a weight, c1+c2=1，LC(i) Representing a set of all next-level said logical nodes connected to node i; c (i) represents a set of all next-level nodes connected to node i; l isijRepresenting the logical operation values of nodes i to j; sijIndicating the hop probability of node i jumping to node j.
7. A three-valued logic question-answering consulting optimization system, comprising:
the node creation module is used for respectively performing semantic analysis and logic analysis on a pre-acquired knowledge base to respectively acquire semantic nodes and logic nodes;
the logic calculation module is used for calculating the three-value logic calculation values of the logic nodes and the semantic nodes according to the pre-acquired inquiry text, and comprises the following steps: extracting key features of the query text; comparing the key features with the semantic nodes and the logic nodes respectively, and judging ternary logic values corresponding to the semantic nodes and the logic nodes respectively according to a preset ternary logic rule; according to the three-valued logic values of the logic nodes, calculating to obtain three-valued logic operation values among the logic nodes by combining the connection relation among the logic nodes; wherein the three-valued logic value comprises: true, false, and unknown;
the probability calculation module is used for counting the skipping probability of the semantic nodes; calculating the probability of each node for obtaining an exact answer according to the jump probability and the three-value logic operation value;
and the optimizing module is used for acquiring the optimal consultation question-answering node according to the probability of acquiring the exact answer by each node.
8. A computer-readable storage medium, in which a computer program is stored which, when loaded and executed by a processor, carries out the method of any one of claims 1 to 6.
9. An apparatus comprising a processor and a memory; wherein the content of the first and second substances,
the memory is used for storing a computer program;
the processor is configured to load and execute the computer program, such that the apparatus performs the method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910674219.2A CN110457455B (en) | 2019-07-25 | 2019-07-25 | Ternary logic question-answer consultation optimization method, system, medium and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910674219.2A CN110457455B (en) | 2019-07-25 | 2019-07-25 | Ternary logic question-answer consultation optimization method, system, medium and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110457455A CN110457455A (en) | 2019-11-15 |
| CN110457455B true CN110457455B (en) | 2022-02-22 |
Family
ID=68483381
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910674219.2A Active CN110457455B (en) | 2019-07-25 | 2019-07-25 | Ternary logic question-answer consultation optimization method, system, medium and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110457455B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111459498A (en) * | 2020-04-09 | 2020-07-28 | 北京理工大学 | Method and device for realizing quick response of policy engine by using logic expression and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107861951A (en) * | 2017-11-17 | 2018-03-30 | 康成投资（中国）有限公司 | Session subject identifying method in intelligent customer service |
| CN109670024A (en) * | 2018-12-17 | 2019-04-23 | 北京百度网讯科技有限公司 | Logical expression determines method, apparatus, equipment and medium |
| CN109902165A (en) * | 2019-03-08 | 2019-06-18 | 中国科学院自动化研究所 | Intelligent interactive answering method, system, device based on Markov Logic Networks |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060053000A1 (en) * | 2004-05-11 | 2006-03-09 | Moldovan Dan I | Natural language question answering system and method utilizing multi-modal logic |
| US10108600B2 (en) * | 2016-09-16 | 2018-10-23 | Entigenlogic Llc | System and method of attribute, entity, and action organization of a data corpora |
-
2019
- 2019-07-25 CN CN201910674219.2A patent/CN110457455B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107861951A (en) * | 2017-11-17 | 2018-03-30 | 康成投资（中国）有限公司 | Session subject identifying method in intelligent customer service |
| CN109670024A (en) * | 2018-12-17 | 2019-04-23 | 北京百度网讯科技有限公司 | Logical expression determines method, apparatus, equipment and medium |
| CN109902165A (en) * | 2019-03-08 | 2019-06-18 | 中国科学院自动化研究所 | Intelligent interactive answering method, system, device based on Markov Logic Networks |
Non-Patent Citations (2)
| Title |
|---|
| Exploring Markov Logic Networks for Question Answering;Tushar Khot;《Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing》;20150921;全文 * |
| 基于深层逻辑知识抽取的问答系统研究;刘昭;《中国优秀硕士学位论文全文数据库 信息科技辑》;20150315(第3期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110457455A (en) | 2019-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11227118B2 (en) | Methods, devices, and systems for constructing intelligent knowledge base | |
| US8457950B1 (en) | System and method for coreference resolution | |
| CN110909725A (en) | Method, device and equipment for recognizing text and storage medium | |
| CN115795061B (en) | Knowledge graph construction method and system based on word vector and dependency syntax | |
| CN112329460A (en) | Text topic clustering method, device, equipment and storage medium | |
| CN110457484B (en) | Method, system, medium and device for logical expression based on graph | |
| CN116304748A (en) | Text similarity calculation method, system, equipment and medium | |
| CN110245700B (en) | Classification model construction method, classification model and object identification method | |
| KR20120047622A (en) | System and method for managing digital contents | |
| CN110457455B (en) | Ternary logic question-answer consultation optimization method, system, medium and equipment | |
| CN113590811A (en) | Text abstract generation method and device, electronic equipment and storage medium | |
| WO2023246849A1 (en) | Feedback data graph generation method and refrigerator | |
| CN112417147A (en) | Method and device for selecting training samples | |
| CN115577147A (en) | Visual information map retrieval method and device, electronic equipment and storage medium | |
| CN112541357B (en) | Entity identification method and device and intelligent equipment | |
| CN111259975B (en) | Method and device for generating classifier and method and device for classifying text | |
| CN112182235A (en) | Method and device for constructing knowledge graph, computer equipment and storage medium | |
| JP2008165572A (en) | Data classification device and data classification program | |
| CN115688771B (en) | Document content comparison performance improving method and system | |
| CN110688446B (en) | Sentence meaning mathematical space representation method, system, medium and equipment | |
| CN115905456B (en) | Data identification method, system, equipment and computer readable storage medium | |
| CN116501875B (en) | Document processing method and system based on natural language and knowledge graph | |
| US20240046039A1 (en) | Method for News Mapping and Apparatus for Performing the Method | |
| US20240070387A1 (en) | Method for Determining News Ticker Related to News Based on Sentence Ticker and Apparatus for Performing the Method | |
| US20240070396A1 (en) | Method for Determining Candidate Company Related to News and Apparatus for Performing the Method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 400000 6-1, 6-2, 6-3, 6-4, building 7, No. 50, Shuangxing Avenue, Biquan street, Bishan District, Chongqing Applicant after: CHONGQING ZHAOGUANG TECHNOLOGY CO.,LTD. Address before: 400000 2-2-1, 109 Fengtian Avenue, tianxingqiao, Shapingba District, Chongqing Applicant before: CHONGQING ZHAOGUANG TECHNOLOGY CO.,LTD. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Optimization method, system, medium and equipment of a ternary logic question answering consultation Effective date of registration: 20221115 Granted publication date: 20220222 Pledgee: Bishan sub branch of Chongqing Three Gorges Bank Co.,Ltd. Pledgor: CHONGQING ZHAOGUANG TECHNOLOGY CO.,LTD. Registration number: Y2022980021313 |