CN111400503A - Knowledge graph generation method based on multiple indexes - Google Patents
Knowledge graph generation method based on multiple indexes Download PDFInfo
- Publication number
- CN111400503A CN111400503A CN202010126582.3A CN202010126582A CN111400503A CN 111400503 A CN111400503 A CN 111400503A CN 202010126582 A CN202010126582 A CN 202010126582A CN 111400503 A CN111400503 A CN 111400503A
- Authority
- CN
- China
- Prior art keywords
- entity
- attribute
- information
- knowledge graph
- confidence
- 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.)
- Granted
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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a knowledge graph generation method based on multiple indexes, which comprises the following steps: and when a new entity appears in the input result, outputting the processing scheme of the entity through three indexes of the editing distance, the text sentence vector and the sound volume. And when a new relation appears in the input result, judging the processing scheme of the relation according to the co-occurrence sound volume and the relation sound volume of the entity corresponding to the relation. And when a new entity attribute appears in the input result, judging the processing scheme of the entity attribute according to the editing distance and the volume of the attribute. And when a new relation attribute appears in the input result, judging the processing scheme of the relation attribute according to the editing distance and the volume of the relation attribute. The method has good effect on the specific field with less data, improves the quality of the knowledge graph in the specific field under the condition of little or no manual participation, has high accuracy of identifying and updating the knowledge graph of the input information, and can improve the accuracy of automatically generating the knowledge graph by knowledge.
Description
Technical Field
The invention relates to the technical field of computer text processing, in particular to a knowledge graph generation method.
Background
Knowledge-graphs are a common knowledge visualization and storage tool. Due to the complexity and diversity of knowledge, making a knowledge graph requires a significant amount of manpower. The knowledge graph generated automatically is generally only used for the field with a large amount of data, and has poor effect on the specific field with less data. There is therefore a need for a method of improving the quality of domain-specific knowledge maps with little or no human involvement.
Disclosure of Invention
Aiming at the problems, the invention provides a knowledge graph generation method based on multiple indexes, which specifically comprises the following steps:
s001, defining a database data structure, wherein the defined data structure comprises an entity, a relation, an entity attribute and a relation attribute; the entity at least comprises three attributes of name, alternative name and document ID; the relation is a directed link between two entities, the link starts from a starting entity, points to an ending entity and at least comprises a name attribute; the entity attribute corresponds to a specific entity and is key value pair information in the corresponding entity; the relationship attribute corresponds to a specific relationship and is key value pair information in the corresponding relationship;
s002, inputting information; the information is one or more of an entity, a relationship, an entity attribute and a relationship attribute;
s003, matching the input information one by one, directly executing the step S007 if the matching is successful, and executing the step S004 if the matching is failed;
s004, information matching: correspondingly generating a processing scheme according to the data structure type of the matching failure information;
s005, calculating the confidence of each processing scheme by using the multi-index parameters;
s006, selecting the processing scheme of the matching failure information according to the confidence;
and S007, updating the data of the database, namely updating the knowledge graph, by using the successfully matched input information or the selected processing scheme, and starting from the step S002 when the knowledge graph is updated by subsequently inputting the information again.
As a further description of the present invention, the entity, the relationship, the entity attribute and the relationship attribute information input in step S002 are obtained by manual labeling or data model prediction.
Furthermore, an information filtering step is further included between step S002 and step S003, filtering is performed through the sound volume parameter of the input information and the set sound volume threshold, and the input information with the sound volume smaller than the sound volume threshold is filtered.
Furthermore, in the information matching in step S003, the processing schemes generated according to the types of the input information are different, and the method for calculating the confidence degrees of the corresponding processing schemes in the subsequent step S005 is also different.
Furthermore, when the input information type is an entity and the information matching fails, the correspondingly generated processing scheme comprises four types, namely fusion into a certain database entity, fusion into a certain new entity, newly-added entity and abandonment; the confidence degrees of the two processing schemes fused into a certain database entity and a certain new entity are calculated by three indexes of editing distance, text sentence vectors and acoustic quantity, and the calculation formula is as follows: the confidence coefficient is (volume index + edit distance index + sentence vector index)/3.
Furthermore, when the input information type is a relationship and the information matching fails, the correspondingly generated processing scheme comprises a newly added relationship and a discarded relationship; the confidence of the newly added relationship processing mode is calculated by two indexes of co-occurrence sound volume and relationship sound volume of the starting entity and the ending entity, and the calculation formula is as follows: confidence coefficient is (co-occurrence sound volume index + sound volume index)/2.
Furthermore, when the input information type is an entity attribute and the information matching fails, the correspondingly generated processing scheme comprises two types, namely modified or newly added attribute and abandonment; the confidence of the processing mode of the modified or newly added attribute is calculated by two indexes of the editing distance and the sound volume of the entity attribute, and the calculation formula is as follows: confidence coefficient is (edit distance index + sound volume index)/2.
Furthermore, when the input information type is the relationship attribute and the information matching fails, the correspondingly generated processing scheme comprises two types of correction or new attribute and abandonment; the confidence of the processing mode of the modified or newly added attribute is calculated by two indexes of the editing distance and the sound volume of the entity attribute, and the calculation formula is as follows: confidence coefficient is (edit distance index + sound volume index)/2.
Further, the manner of selecting the processing scheme in step S006 includes manual selection and automatic machine execution selection.
Further, the machine automatically performs the selection including inputting a confidence threshold, and automatically performing the processing scheme with the highest confidence when the scheme with the highest confidence among all the processing schemes in all the categories is larger than the confidence threshold, otherwise selecting to be discarded.
The invention has the beneficial effects that:
the knowledge graph generation method based on multiple indexes has good effect on specific fields with less data, improves the quality of the knowledge graph in the specific fields under the condition of little or no manual participation, has high accuracy of identifying input information and updating the knowledge graph, can improve the accuracy of automatically generating the knowledge graph by knowledge, and can reduce the manual workload when manual intervention is needed.
Drawings
FIG. 1 is an overall flow diagram of the process of the present invention;
FIG. 2 is an exemplary knowledge graph database structure according to the present invention;
FIG. 3 is a flow chart of a method of knowledge-graph entity generation in accordance with the present invention;
FIG. 4 is a flow chart of a knowledge-graph relationship generation method of the present invention;
FIG. 5 is a flowchart of a method for knowledge-graph entity attribute generation in accordance with the present invention;
FIG. 6 is a flow chart of a method for knowledge-graph relationship attribute generation in accordance with the present invention.
Detailed Description
The following detailed description of specific embodiments of the invention, taken in conjunction with the accompanying drawings, will make apparent that the described embodiments are only some, but not all embodiments of the invention.
Fig. 1 shows an overall flowchart of a knowledge graph generation method based on multiple indexes, which includes the following steps:
and S001, defining a database data structure. The database may be a null database or a non-null database. If the database is not empty, each entity in the database should contain at least three attributes, namely name, alternative name and document ID. The name is the name which can represent the entity most, and is called the name of the entity except the name, and the document ID is the ID list of the document which appears in the entity. A relationship refers to a directed link between two entities, the link starting from a starting entity and pointing to an ending entity. The relationship minimally contains a name attribute. Fig. 2 is an example of this data structure.
And S002, inputting the entity, the relation, the entity attribute and the relation attribute information.
The relationship must correspond to a specific starting entity and a specific ending entity, the entity attribute must correspond to a specific entity, and the relationship attribute must correspond to a specific relationship.
The information may be derived from manual sorting or from algorithm recognition results.
Wherein the input information should contain a list of IDs of documents in which the information is located.
Table 1 is a specific example of inputting entities, relationships, entity attributes, and relationship attribute information in this embodiment.
Table 1 inputs samples:
and S003, filtering information with too low sound volume. The volume is the number of times that the entity, relationship, entity attribute or relationship attribute appears in the original document. And respectively inputting the sound volume thresholds of the entity, the relation, the entity attribute and the relation attribute, and filtering the input information of which the sound volume is less than the sound volume threshold without any treatment.
And S004, generating an information matching and processing scheme, wherein the information matching and processing scheme comprises four matching processing flows including an entity, a relationship, an entity attribute and a relationship attribute, and flow charts of the four matching processing flows correspond to those shown in the attached figures 3-6 respectively. Specifically, the method comprises the following steps:
the entity matching processing flow comprises entity matching and entity processing scheme generation.
Entity matching: all input entities are processed one by one. First, it is confirmed whether the entity exists in the database. The specific method is to search the database, if the name or the alternative name of only one entity in the database is equal to the input entity, the matching is successful, the entity exists in the database, otherwise, the entity does not exist in the database, and the input entity is a new entity.
For the input of table 1 and the database in fig. 2, the entity "mingming" match was successful and the entity "mingming" did not.
And if the input entity is successfully matched, adding the document ID of the input entity into the document ID of the entity corresponding to the database.
And if the input entity fails to be matched, the input entity is a new entity, and the next operation is carried out.
And (3) generating an entity processing scheme: for new entities, there are 4 types of processing schemes, which are: merging into a certain database entity, merging into a certain new entity, adding new entities and discarding. The first two types of schemes may contain more than one treatment scheme. The different processing schemes need to be calculated separately.
Wherein, such a scheme requires traversing all database entities, calculating the confidence of fusion into them.
The confidence index is composed of the average value of three indexes
Confidence coefficient (sound volume index + edit distance index + sentence vector index)/3 (1)
In the formula (1), the sound volume index, the edit distance index, and the sentence vector index are calculated in the following manner:
volume index (2) input entity volume/total number of documents
In formula (2), the total number of documents is the total number of documents involved in this information input.
In formula (3), the names and alternative names of the database entities are respectively calculated, and the maximum value is taken as an edit distance index
Sentence vector index (text sentence vector of input entity and text sentence vector of database entity) (4)
In the formula (4), the text where the entity is located can be found through the text ID, the texts are combined, and then the sentence vector is calculated, so that the result of the formula (4) can be obtained.
Where "fused into a new entity," such a scheme requires traversing all known new entities, computing the confidence into which to fuse. The method of calculating confidence is consistent with the "fused into some database entity" class of methods.
Wherein, the scheme only comprises a processing scheme, in order to add an entity with the name of the input entity as the name in the database, the confidence coefficient of the scheme is 1- ('fused to the entity of a certain database' the maximum value of the confidence coefficient in the scheme)
Where "obsolete," such methods include only one processing scheme that does not output confidence in order to discard the input entity.
The relationship matching processing flow comprises relationship matching and relationship processing scheme generation.
And (3) relation matching: and processing all input relations one by one, firstly, confirming whether the relation exists in the database, if the initial entity and the end entity of the input relation are successfully matched, and simultaneously, the relation exists between the initial entity and the end entity in the database, indicating that the relation exists in the database, otherwise, the relation does not exist in the database, and the input relation is a new relation.
For the input of table 1 and the database in fig. 2, the relationship "brother" from xiaoming to xiaohong was successfully matched and the relationship "friend" from mingming to xiaolan was not successfully matched.
And if the input relation matching fails, the input relation is a new relation, and the next operation is carried out.
The relational processing scheme generates: for the new relationship, there are 2 types of processing schemes, which are respectively a new relationship and a discard relationship.
Wherein, the confidence index of the scheme is composed of two index average values:
confidence coefficient (co-occurrence sound volume index + sound volume index)/2 (5)
In the formula (5), the co-occurrence loudness indicator is calculated in the following manner:
the acoustic quantity index is calculated in the following manner
Where "obsolete," such methods include only one processing scheme that does not output confidence in order to discard the input entity.
The entity attribute matching processing flow comprises entity attribute matching and entity attribute processing scheme generation.
And (3) entity attribute matching: and processing all input entity attributes one by one, firstly, determining whether the entity attributes exist in the database, if the entity matching corresponding to the input entity attributes is successful and the entity in the database has the attributes, the matching is successful, otherwise, the matching is failed.
Wherein, the attribute is a key value pair composed of two parts, namely an attribute key and a value.
For the input of Table 1 and the database in FIG. 2, the entity attribute "height: 170cm "failed the match.
And if the matching fails, the input entity attribute is a new entity attribute, and the next operation is carried out.
And generating an entity attribute processing scheme: the new entity attributes of the chess have two types of processing schemes, namely modified or newly added attributes and obsolete.
Wherein, the confidence index of the scheme is composed of two indexes of average value:
confidence coefficient (edit distance index + sound volume index)/2 (8)
In equation (8), the edit distance index is calculated as follows:
in the formula (9), entities with the same attribute key in the database are found, calculation is performed respectively, and the maximum value is taken as the edit distance index.
The acoustic quantity index is calculated in the following manner
Where "obsolete," such methods only include a processing scheme that does not output confidence in order to discard the input entity attribute.
The relation attribute matching processing flow comprises relation attribute matching and relation attribute processing scheme generation.
And (3) matching the relationship attributes: and processing all the input relation attributes one by one, firstly, determining whether the relation attribute exists in the database, if the input relation attribute corresponding relation is successfully matched and the relation in the database has the attribute, the matching is successful, otherwise, the matching is failed.
For the input of table 1 and the database in fig. 2, the entity attribute "affinity: low "match fails.
And if the matching fails, the input relationship attribute is a new relationship attribute, and the next operation is carried out.
Generating a relationship attribute processing scheme: for the new relationship attributes, there are two types of processing schemes, namely, modified or added attributes and obsolete.
Wherein, the confidence index of the scheme is composed of two indexes of average value:
confidence coefficient (edit distance index + sound volume index)/2 (8)
In equation (8), the edit distance index is calculated as follows:
in the formula (9), the relationship of the same attribute key in the database is found, and the calculation is performed respectively, and the maximum value is taken as the edit distance index.
The acoustic quantity index is calculated in the following manner
Among these, "discard", this type of method has only one processing scheme that does not output confidence in order to discard the input relationship attribute.
And S005, processing scheme selection, including two modes of manual selection and automatic machine execution selection. If manual processing is selected, each input scheme is selected according to the confidence level of the scheme and personal experience. If the scheme with the highest confidence coefficient is automatically executed by the machine, a confidence coefficient threshold value is input, if the scheme with the highest confidence coefficient is larger than the confidence coefficient threshold value in all the types of processing schemes, the processing scheme with the highest confidence coefficient is automatically executed, and if not, the processing scheme is selected to be discarded.
Wherein, the selection must be performed according to the order of the entity, the relationship, the entity attribute, and the relationship attribute.
In the process of selecting the processing scheme, when the processing scheme of the starting entity or the ending entity corresponding to the relationship is selected to be fused to a certain entity, the starting entity or the ending entity corresponding to the relationship is changed accordingly. When the entity corresponding to the entity attribute is selected to be fused to a certain entity, the entity corresponding to the entity attribute changes accordingly. When the relationship corresponding to the relationship attribute is selected to be fused to a certain relationship, the relationship corresponding to the relationship attribute is changed.
And S006, updating the knowledge graph, and modifying the knowledge graph according to the processing scheme. When the input entity is selected to be fused to a certain entity scheme, the name of the input entity is added to the nickname attribute of the database entity, and the document ID of the input entity is added to the document ID attribute. Entity information is entered at the database new entity when an input entity is selected to the new entity scheme. When the input relationship is selected to the new relationship scheme, the relationship is created in the database. When the input entity attribute is selected to modify the current attribute scheme, the attribute is modified or newly established in the corresponding entity of the database. When the input relation attribute is selected to modify the current attribute scheme, the attribute is modified or newly established in the corresponding relation of the database.
The foregoing is illustrative of the preferred embodiments of the present invention only and is not to be construed as limiting the claims. The invention is not limited to the above embodiments, the specific construction of which allows variations, and in any case variations, which are within the scope of the invention as defined in the independent claims.
Claims (10)
1. A knowledge graph generation method based on multiple indexes is characterized by comprising the following steps:
s001, defining a database data structure, wherein the defined data structure comprises an entity, a relation, an entity attribute and a relation attribute; the entity at least comprises three attributes of name, alternative name and document ID; the relation is a directed link between two entities, the link starts from a starting entity, points to an ending entity and at least comprises a name attribute; the entity attribute corresponds to a specific entity and is key value pair information in the corresponding entity; the relationship attribute corresponds to a specific relationship and is key value pair information in the corresponding relationship;
s002, inputting information; the information is one or more of an entity, a relationship, an entity attribute and a relationship attribute;
s003, matching the input information one by one, directly executing the step S007 if the matching is successful, and executing the step S004 if the matching is failed;
s004, information matching: correspondingly generating a processing scheme according to the data structure type of the matching failure information;
s005, calculating the confidence of each processing scheme by using the multi-index parameters;
s006, selecting the processing scheme of the matching failure information according to the confidence;
and S007, updating the data of the database, namely updating the knowledge graph, by using the successfully matched input information or the selected processing scheme, and starting from the step S002 when the knowledge graph is updated by subsequently inputting the information again.
2. The multi-index based knowledge graph generating method according to claim 1, wherein: the information input in step S002 is obtained by manual labeling or data model prediction.
3. The multi-index based knowledge graph generating method according to claim 1, wherein: and an information filtering step is further included between the step S002 and the step S003, and the input information with the sound volume smaller than the sound volume threshold is filtered through the sound volume parameter of the input information and the set sound volume threshold.
4. The multi-index based knowledge graph generating method according to claim 1, wherein: in the information matching in step S003, the processing schemes generated correspondingly according to the types of the input information are different, and the method for calculating the confidence degrees of the corresponding processing schemes in the subsequent step S005 is also different.
5. The multi-index based knowledge graph generation method according to claim 4, wherein: when the input information type is an entity and the information matching fails, the correspondingly generated processing scheme comprises four types, namely integration into a certain database entity, integration into a certain new entity, addition of a new entity and abandonment; the confidence degrees of the two processing schemes fused into a certain database entity and a certain new entity are calculated by three indexes of editing distance, text sentence vectors and acoustic quantity, and the calculation formula is as follows: the confidence coefficient is (volume index + edit distance index + sentence vector index)/3.
6. The multi-index based knowledge graph generation method according to claim 4, wherein: when the input information type is a relationship and the information matching fails, the correspondingly generated processing scheme comprises a newly added relationship and a waste relationship; the confidence of the newly added relationship processing mode is calculated by two indexes of co-occurrence sound volume and relationship sound volume of the starting entity and the ending entity, and the calculation formula is as follows: confidence coefficient is (co-occurrence sound volume index + sound volume index)/2.
7. The multi-index based knowledge graph generation method according to claim 4, wherein: when the input information type is an entity attribute and the information matching fails, the correspondingly generated processing scheme comprises two types of correction or new attribute addition and abandonment; the confidence of the processing mode of the modified or newly added attribute is calculated by two indexes of the editing distance and the sound volume of the entity attribute, and the calculation formula is as follows: confidence coefficient is (edit distance index + sound volume index)/2.
8. The multi-index based knowledge graph generation method according to claim 4, wherein: when the input information type is a relational attribute and the information matching fails, the correspondingly generated processing scheme comprises two types of correction or new attribute addition and abandonment; the confidence of the processing mode of the modified or newly added attribute is calculated by two indexes of the editing distance and the sound volume of the entity attribute, and the calculation formula is as follows: confidence coefficient is (edit distance index + sound volume index)/2.
9. The multi-index based knowledge graph generation method according to claim 4, wherein: the manner of selecting the processing scheme in step S006 includes manual selection and automatic machine execution selection.
10. The multi-index based knowledge graph generating method according to claim 9, wherein: the machine automatically performs selection, wherein a confidence threshold value is input, and when the scheme with the maximum confidence in all the processing schemes in all the categories is larger than the confidence threshold value, the processing scheme with the maximum confidence is automatically performed, otherwise, the processing scheme is selected to be abandoned.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010126582.3A CN111400503B (en) | 2020-02-28 | 2020-02-28 | Knowledge graph generation method based on multiple indexes |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010126582.3A CN111400503B (en) | 2020-02-28 | 2020-02-28 | Knowledge graph generation method based on multiple indexes |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111400503A true CN111400503A (en) | 2020-07-10 |
| CN111400503B CN111400503B (en) | 2023-09-22 |
Family
ID=71428498
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010126582.3A Active CN111400503B (en) | 2020-02-28 | 2020-02-28 | Knowledge graph generation method based on multiple indexes |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111400503B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106355627A (en) * | 2015-07-16 | 2017-01-25 | 中国石油化工股份有限公司 | Method and system used for generating knowledge graphs |
| CN107391677A (en) * | 2017-07-21 | 2017-11-24 | 深圳狗尾草智能科技有限公司 | Carry the generation method and device of the Universal Chinese character knowledge mapping of entity-relationship-attribute |
-
2020
- 2020-02-28 CN CN202010126582.3A patent/CN111400503B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106355627A (en) * | 2015-07-16 | 2017-01-25 | 中国石油化工股份有限公司 | Method and system used for generating knowledge graphs |
| CN107391677A (en) * | 2017-07-21 | 2017-11-24 | 深圳狗尾草智能科技有限公司 | Carry the generation method and device of the Universal Chinese character knowledge mapping of entity-relationship-attribute |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111400503B (en) | 2023-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105608218B (en) | The method for building up of intelligent answer knowledge base establishes device and establishes system | |
| US10515155B2 (en) | Conversational agent | |
| WO2018040899A1 (en) | Error correction method and device for search term | |
| CN111090461B (en) | Code annotation generation method based on machine translation model | |
| JP4429236B2 (en) | Classification rule creation support method | |
| CN102867040B (en) | Chinese search engine mixed speech-oriented query error correction method and system | |
| CN112232925A (en) | Method for carrying out personalized recommendation on commodities by fusing knowledge maps | |
| CN113361266B (en) | Text error correction method, electronic device and storage medium | |
| CN110555101A (en) | customer service knowledge base updating method, device, equipment and storage medium | |
| CN112560450B (en) | Text error correction method and device | |
| CN110377739A (en) | Text sentiment classification method, readable storage medium storing program for executing and electronic equipment | |
| CN108182182B (en) | Method and device for matching documents in translation database and computer readable storage medium | |
| CN104485106A (en) | Voice recognition method, voice recognition system and voice recognition equipment | |
| EP3525107A1 (en) | Conversational agent | |
| CN115982337A (en) | Intention identification method for transaction handling in online conversation | |
| CN112181949A (en) | Online data modeling method and device | |
| CN111400503B (en) | Knowledge graph generation method based on multiple indexes | |
| CN110032642B (en) | Modeling method of manifold topic model based on word embedding | |
| Moraglio et al. | Evolving recursive programs using non-recursive scaffolding | |
| CN112632910A (en) | Operation encoding method, electronic device and storage device | |
| WO2022156450A1 (en) | Knowledge base query method and apparatus, computer device, and storage medium | |
| JP6474873B1 (en) | Class token correspondence determination apparatus and method, computer program therefor, and storage medium storing the computer program | |
| CN111222188A (en) | Method for analyzing and mining style preference of client through IM chat content | |
| CN111310066A (en) | Friend recommendation method and system based on topic model and association rule algorithm | |
| CN110147393A (en) | The entity resolution method in data-oriented space |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |