CN106126503B - Service field positioning method and terminal - Google Patents
Service field positioning method and terminal Download PDFInfo
- Publication number
- CN106126503B CN106126503B CN201610545704.6A CN201610545704A CN106126503B CN 106126503 B CN106126503 B CN 106126503B CN 201610545704 A CN201610545704 A CN 201610545704A CN 106126503 B CN106126503 B CN 106126503B
- Authority
- CN
- China
- Prior art keywords
- knowledge
- knowledge graph
- distance
- determining
- keywords
- 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
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a business field positioning method and a terminal, which extracts keywords from interactive sentences input by a user; determining a knowledge graph to which the keyword belongs, then determining the distance of each keyword in the interactive sentence in each knowledge graph of at least two knowledge graphs, then determining the minimum value from the distances, and finally determining the business field corresponding to the interactive sentence according to the knowledge graph corresponding to the minimum value. In the process, the distance of the keywords in the knowledge graph indicates the distance of each keyword relation, namely the intimacy degree, so that the service field of the user tendency can be determined according to the minimum distance, the purpose of improving the positioning accuracy of the service field is realized, multiple rules and weights are written for the labels in the service field positioning process, and the complexity of the service field positioning is reduced.
Description
Technical Field
The embodiment of the invention relates to a service field positioning technology, in particular to a service field positioning method and a terminal.
Background
At present, there are multiple fields of inquiry in human-computer interaction, for example, a user interacts with a terminal through voice, keyboard operation, screen operation, etc., so as to inquire about services such as movie content, music content, weather information, date information, stock information, encyclopedia information, etc. In the process, the terminal accurately positions the intention of the user in a short time and returns a correct result, and the terminal is very important in man-machine interaction.
In order to quickly and accurately locate the intention of a user, in the existing scheme, a label is set for a word of a labeled service, multiple rules and weights are compiled for the label, and then the service is located according to the rules and weights of the label.
In the service positioning process, multiple rules and weights need to be compiled for the tags, the process is complex, and the accuracy of service positioning is low.
Disclosure of Invention
The invention provides a service field positioning method and a terminal, which achieve the purpose of improving the service positioning accuracy.
In a first aspect, an embodiment of the present invention provides a service location method, including:
receiving an interactive statement input by a user, and acquiring a keyword in the interactive statement;
determining a knowledge graph to which the keyword belongs, wherein the knowledge graph indicates the service field of the keyword, the number of the keyword is at least two, and the number of the knowledge graph is at least two;
for each of the at least two knowledge graphs, determining a distance of the interactive sentence in the knowledge graph, wherein the distance is the sum of distances of keywords of the interactive sentence in the knowledge graph;
determining a minimum value from the distances;
and determining the service field corresponding to the interactive statement according to the knowledge graph corresponding to the minimum value.
In a second aspect, an embodiment of the present invention provides a service location apparatus, including:
the receiving module is used for receiving interactive sentences input by a user and acquiring keywords in the interactive sentences;
the knowledge graph determining module is used for determining a knowledge graph to which the keyword belongs, wherein the knowledge graph indicates the service field of the keyword, the number of the keyword is at least two, and the number of the knowledge graph is at least two;
a distance determining module, configured to determine, for each of the at least two knowledge graphs, a distance of the interactive sentence within the knowledge graph, where the distance is a sum of distances of keywords of the interactive sentence within the knowledge graph;
a minimum value determining module for determining a minimum value from the distances;
and the service field determining module is used for determining the service field corresponding to the interactive statement according to the knowledge graph corresponding to the minimum value.
The business field positioning method and the terminal provided by the invention extract keywords from interactive sentences input by a user; determining a knowledge graph to which the keyword belongs, then determining the distance of each keyword in the interactive sentence in each knowledge graph of at least two knowledge graphs, then determining the minimum value from the distances, and finally determining the business field corresponding to the interactive sentence according to the knowledge graph corresponding to the minimum value. In the process, the distance of the keywords in the knowledge graph indicates the distance of each keyword relation, namely the intimacy degree, so that the service field of the user tendency can be determined according to the minimum distance, the purpose of improving the positioning accuracy of the service field is realized, multiple rules and weights are written for the labels in the service field positioning process, and the complexity of the service field positioning is reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the method of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the method of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive effort.
Fig. 1 is a process diagram of a service domain location method of the present invention;
FIG. 2 is a flowchart of a first embodiment of a service area location method of the present invention;
FIG. 3 is a diagram illustrating an example of a knowledge graph in the service domain location method of the present invention;
FIG. 4A is a diagram illustrating an example of keywords in a knowledge graph according to the method for locating a business segment of the present invention;
FIG. 4B is a diagram illustrating keywords in another knowledge graph according to the method for locating a business segment of the present invention;
FIG. 5 is an exemplary diagram of a generic knowledge graph in the service domain location method of the present invention;
FIG. 6A is a diagram illustrating an example of a video knowledge graph in the service domain positioning method of the present invention;
FIG. 6B is a diagram illustrating an example of a music knowledge base in the method for domain location according to the present invention;
fig. 7 is a schematic structural diagram of a first embodiment of the terminal according to the present invention;
fig. 8 is a schematic structural diagram of a second terminal embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a process diagram of a service domain positioning method according to the present invention. Referring to fig. 1, in the present embodiment, first, a user inputs, for example, through interaction with a terminal through voice input, keyboard operation, screen operation, and the like, and inputs an interactive language for querying services such as movie content, music content, weather information, date information, stock information, encyclopedia information, and the like; then, the terminal carries out service field positioning according to the user input; and finally, the terminal outputs a result, namely, the service field corresponding to the interactive statement is determined. In the following, on the basis of fig. 1, a service location method of the present invention is described in detail, and specifically, refer to fig. 2.
Fig. 2 is a flowchart of a first embodiment of a service domain positioning method according to the present invention. The invention is suitable for scenes needing to accurately and quickly position the service field. Specifically, the present embodiment includes:
101. receiving an interactive statement input by a user, and acquiring a keyword in the interactive statement.
In the embodiment of the invention, in the man-machine interaction process, a user sends an instruction to the terminal through an interactive statement, for example, the instruction interacts with the terminal through voice input, keyboard operation, screen operation and the like, so that businesses such as movie content, music content, weather information, date information, stock information, encyclopedia information and the like are inquired. In this step, after receiving the interactive language, the terminal extracts keywords, such as words with a definite domain tendency, from the interactive sentence. For example, when a user queries a video service of the dune super lead actor, the user inputs "dune super", and the extracted keyword is "dune super"; for another example, when the user inquires about the songs of liu de hua, the user inputs "ice rain of liu de hua", and the extracted keywords are "liu de hua" and "ice rain", and in addition, there is a non-keyword "for" liu de hua ".
102. Determining a knowledge graph to which the keyword belongs, wherein the knowledge graph indicates the service field of the keyword, the number of the keyword is at least two, and the number of the knowledge graph is at least two.
The knowledge graph is essentially a semantic network, and the storage form of the knowledge graph is the interactive connection of nodes and edges, wherein the nodes represent entity words or concept words, and the edges represent various semantic relationships between entity words and entity words, between concept words and concept words, and between entity words and concept words, so as to reveal the relationships and distances between entity words and entity words, between concept words and concept words, and between entity words and concept words, and the entity words and concept words are the keywords described in the embodiment of the present invention. The distance is understood to be the degree of closeness.
In this step, when determining the knowledge graph to which the keyword belongs, the determination may be performed by looking up a table or the like.
In a possible implementation, the terminal maintains a relation table indicating the knowledge graph of the keywords. For example, if the keyword is "liu de hua", then "liu de hua" indicated in the relationship table belongs to the movie knowledge graph and the music knowledge graph, and is recorded as liu de hua: video/music.
In another possible implementation, the terminal maintains a generic knowledge-graph that is used primarily to identify which keywords belong to which knowledge-graph (similar to the lookup table described above). In addition, since the input language of the user also includes some words without definite domain tendency, it is referred to as non-keyword hereinafter. In order to realize that all words contained in the interactive sentences can find corresponding places in the general knowledge graph, in the embodiment of the invention, the general knowledge graph also maintains some non-keywords, the non-keywords are used for modifying keywords and comprise limiting words, such as 'today', 'tomorrow', 'last year' and the like, relation words, such as 'old official', 'it', 'but' and the like, function words, such as 'latest', and the like, and common general words, such as 'good', 'I', 'Do', and the like, and the non-keywords do not belong to knowledge graphs in any fields.
103. For each of the at least two knowledge-graphs, determining a distance of the interactive sentence within the knowledge-graph, wherein the distance is a sum of distances of keywords of the interactive sentence within the knowledge-graph.
In this step, when there are two keywords extracted from the interactive sentence, the distance between the two keywords may be understood, and when there are more than two keywords extracted from the interactive sentence, the sum of the distances between all the keywords is referred to.
In a possible implementation manner, when each keyword simultaneously belongs to one knowledge graph of at least two knowledge graphs, determining the sum of the distances of the keywords in the knowledge graph according to the connection relation between the keywords in the knowledge graph, wherein the distances are finite values; and then, determining the distance of the interactive sentences in the knowledge graph according to the sum of the distances of the keywords in the knowledge graph.
In another possible implementation manner, when each keyword belongs to different knowledge graphs in at least two knowledge graphs respectively, for each knowledge graph in at least two knowledge graphs, determining that the distance between the keyword belonging to the knowledge graph and the keyword belonging to other knowledge graphs is infinite; and determining that the distance of the interactive statement in the knowledge graph is infinite, wherein the distance is infinite, and the distance is the sum of the distances of the keywords of the interactive statement in the knowledge graph.
Specifically, in the embodiment of the present invention, a knowledge graph represents a service field, the knowledge graph indicates a distance between at least two keywords, the distance between the keywords can be determined according to a connection relationship between the keywords, and when there is a connection relationship, the distance is a finite value. Generally, each keyword contained in the same knowledge graph indicates that the relationship of the keywords is very close and the intimacy degree is higher if the keywords are directly connected, and indicates that the keywords are far if the keywords are indirectly connected, and the distance corresponding to the direct connection relationship is smaller than the distance corresponding to the indirect connection relationship; in different knowledge graphs, the distance between the keywords contained in one knowledge graph and the keywords contained in another knowledge graph can be understood as infinite, that is, there is no relation, and in this case, the distance is infinite.
For example, suppose that a keyword extracted from an interactive sentence input by a user is "dun super", a character relationship knowledge graph of dun super is shown in fig. 3, and fig. 3 is an exemplary schematic diagram of a knowledge graph in the service domain localization method of the present invention.
Referring to fig. 3, the keywords included in the knowledge graph include dun super, grandli, sandui, shu baimei, etc., wherein for the keyword dun super, dun super and grandli are in spouse relationship, dun super and grandli are in friend relationship, dun super and grand bai, sand overflow and invar are in friend relationship, dun super and willow rock have been dug up, the former girlful of dun super is neuning, dun super and dune are in partner relationship, and dun super is directly connected with the characters, which is close in relationship, i.e., higher intimacy degree. Therefore, if "dun superthium" is queried in the movie entertainment knowledge base, the distance can be understood as "1" because the two people are directly connected, and the smaller the distance, the higher the intimacy.
Similarly, in other fields, two keywords are directly connected, which indicates that there is a significant correlation between the two keywords. For example, in a weather knowledge graph, "raining" and "opening umbrella" both belong to the weather knowledge graph, and a causal relationship exists, and the distance can be understood as "1"; for another example, the "snow" and the "down jacket" belong to the weather knowledge map, and can be indirectly connected through a certain relationship of sensory words such as "cold", and the distance can be understood as "2". For another example, the "Liu De Hua" and the "wadded jacket" do not belong to the knowledge graph in the same field, and the distance is infinite, which can be understood as a large distance.
For another example, 5 keywords are extracted from the interactive sentences, and it is assumed that the 5 keywords belong to two knowledge maps at the same time, specifically, refer to fig. 4A and 4B, where fig. 4A is an exemplary schematic diagram of a keyword in one knowledge map in the service domain positioning method of the present invention, and fig. 4B is an exemplary schematic diagram of a keyword in another knowledge map in the service domain positioning method of the present invention.
Referring to fig. 4A and 4B, 5 keywords are shown as black solid dots in the figure. In fig. 4A, the five keywords are connected with each other, and the distance between two business keywords is 1, which indicates that the relation between the five keywords is close and the intimacy degree is high in the knowledge graph shown in fig. 4A; in fig. 4B, the five keywords are distributed and not directly connected to each other, which illustrates that the relationship of the five keywords is far and the intimacy degree is low in the knowledge graph shown in fig. 4B. By comparing fig. 4A and fig. 4B, it is described that the intention of the user is in fig. 4A, the service domain corresponding to the knowledge graph shown in fig. 4A is the service domain in which the user is interested, and the terminal determines that the service domain corresponding to fig. 4A is the service domain corresponding to the interactive statements, and then returns to the service in the service domain shown in fig. 4A.
In addition, assuming that 10 keywords are extracted from the interactive sentence, wherein 5 keywords belong to fig. 4A, and the other 5 keywords belong to fig. 4B, the distance between the keyword in fig. 4A and the keyword in fig. 4B is infinite, the distance between the keywords in fig. 4A is finite, the distance between the keywords in fig. 4B having the direct connection relationship or the indirect connection relationship is finite, and the distance between the keywords having no connection relationship is infinite. Since the distance corresponding to the wireless value is greater than the distance corresponding to the finite value, it is determined that the service domain corresponding to fig. 4A is the service domain corresponding to the interactive statement. If the distribution of the keywords in fig. 4A is as shown in fig. 4B, it is shown that, for each knowledge graph, the distance of each keyword in the knowledge graph is infinite, and at this time, the business field corresponding to the interactive sentence is determined according to the priority of the knowledge graph.
104. A minimum value is determined from the distances.
In the embodiment of the invention, the distance corresponding to the direct connection relation is smaller than the distance corresponding to the indirect connection relation, and the distance with the connection relation is smaller than the distance corresponding to the infinite value.
105. And determining the service field corresponding to the interactive statement according to the knowledge graph corresponding to the minimum value.
In this step, the service field corresponding to the interactive sentence is determined according to the knowledge graph corresponding to the distance of the minimum value.
The business field positioning method provided by the embodiment of the invention extracts keywords from interactive sentences input by a user; determining a knowledge graph to which the keyword belongs, then determining the distance of each keyword in the interactive sentence in each knowledge graph of at least two knowledge graphs, then determining the minimum value from the distances, and finally determining the business field corresponding to the interactive sentence according to the knowledge graph corresponding to the minimum value. In the process, the distance of the keywords in the knowledge graph indicates the distance of each keyword relation, namely the intimacy degree, so that the service field of the user tendency can be determined according to the minimum distance, the purpose of improving the positioning accuracy of the service field is realized, multiple rules and weights are written for the labels in the service field positioning process, and the complexity of the service field positioning is reduced.
Optionally, in the above embodiment, when the interactive statements include non-keywords, in the embodiment of the present invention, the non-keywords are extracted from the interactive statements, and the non-keywords are used to modify the keywords; reducing the business in the business field corresponding to the interactive statement according to the non-keyword; and returning the service within the reduced service range. For example, if the interactive sentence input by the user is "movie in Liu De Hua 2016", then after the movie whose business area is Liu De Hua lead actor is determined, the movie in Liu De Hua performance in 2016 is determined from the plurality of movies by further analyzing the non-keyword, i.e., the restriction word 2016.
In the above embodiments, the present invention has been described in detail with respect to at least two keywords, but the present invention is also applicable to the case where one keyword is used. Specifically, when there is one keyword in the interactive sentence, another keyword having the smallest distance to the keyword is determined from the knowledge graph, and the service corresponding to the keyword having the smallest distance to the keyword is returned. For example, the knowledge graph is only one, and taking fig. 3 as an example, when the keyword is dun timeout, when the relationship of the character is returned, all characters directly connected to dun timeout are returned to the user. For another example, the extracted service keyword is "cloudy day", and the "cloudy day" belongs to a music map and a weather map, and at this time, if the priority of the default music map is higher than that of the weather map, it is determined that the service field is music search. In addition, the user may be asked for further intent through multiple rounds of interaction.
In addition, when the number of the keywords is at least two and the distance values of the at least two keywords corresponding to the knowledge graphs are the same, the service field corresponding to the interactive statement can also be determined according to the priority of the knowledge graphs. Specifically, when the values of the distances corresponding to the knowledge graphs are the same, the knowledge graph with the highest service priority is determined from the at least two knowledge graphs; and determining the service field corresponding to the interactive statement according to the knowledge graph with the highest service priority. For example, referring to fig. 4A and 4B again, assuming that 5 keywords are extracted from the interactive sentences, the closeness of the five keywords in fig. 4A is the same as the closeness of the five keywords in fig. 4B, and at this time, a higher priority knowledge graph is further determined from fig. 4A and 4B; and determining the service field corresponding to the interactive sentence according to the knowledge graph with the highest service priority.
The service location method of the present invention is explained in detail with a specific example.
Specifically, assuming that the input language of the user is "liu de hua sleet", the extracted keywords are "liu de hua", "sleet", and non-keywords, i.e., structural auxiliary words. And determining the knowledge graph to which each keyword belongs through the general knowledge graph. Specifically, referring to fig. 5, fig. 5 is an exemplary schematic diagram of a generic knowledge graph in the service domain positioning method of the present invention.
Referring to fig. 5, the general knowledge graph indicates that the knowledge graph to which liu de hua belongs is video/music, namely, liu de hua is both a singer and an actor, the general knowledge graph indicates that the knowledge graph to which ice rain belongs is video/music/weather, namely, ice rain is a song name and also a korean movie name, and is also a different name of "sleet", belongs to a weather (weather) knowledge graph, and the general knowledge graph indicates that "the knowledge graph does not belong to any knowledge graph, namely, the knowledge graph to which the ice rain belongs is" null ". For the weather knowledge graph, as Liu De Hua does not belong to the weather knowledge graph, the distance between the keyword ice rain and the keyword Liu De Hua in the weather knowledge graph is infinite; in the video knowledge graph, a connection relation exists between the key words Liu Dehua and the key words ice rain, in the music knowledge graph, a connection relation also exists between the key words Liu Dehua and the key words ice rain, and the distance corresponding to the connection relation is smaller than the distance corresponding to the infinite value, so that the distance of the key words in the weather knowledge graph is the largest, and the weather knowledge graph can be ignored. At this time, the video knowledge graph and the music knowledge graph are inquired, and the distance between Liudebua and ice rain in the two knowledge graphs is judged. Specifically, see fig. 6A and 6B.
Fig. 6A is an exemplary schematic diagram of a movie knowledge domain in the service domain positioning method of the present invention. Referring to fig. 6A, the movie knowledge graph includes a plurality of keywords, and the keywords include keywords extracted from the interactive sentences input by the user: liu de hua and sleet (as shown by the solid filled portion in the figure), examples of actors include liu de hua and song cheng constitution, examples of power sources include sleet, song cheng constitution has performed sleet, and sleet geographically belongs to korea, i.e., a korean movie. Assuming that the distance between two directly connected service keywords is 1, the distance between Liudebua and ice rain is 3.
Fig. 6B is an exemplary diagram of a music knowledge graph in the service domain positioning method of the present invention. Referring to fig. 6B, the music knowledge graph includes a plurality of keywords, and the keywords include keywords extracted from interactive sentences input by the user: liu De Hua and ice rain (shown as a solid filled portion in the figure), examples of singers include Liu De Hua, examples of songs include ice rain, and Liu De Hua is a singer of ice rain. Assuming that the distance between two directly connected keywords is 1, the distance between Liudebua and ice rain is 1.
As can be seen from fig. 6A and 6B, in fig. 6B, the distance between liu de hua and ice rain is far smaller than that in fig. 6A, which means that fig. 6B correctly expresses the intention of the user, and therefore, according to fig. 6B, it is determined that the business area corresponding to the interactive sentence is music search.
Similarly, suppose the input language of the user is 'the sky is cloudy', the non-key word, namely the general word 'is' left, and 'the sky' and 'the cloudy' are analyzed. The following can be known through a general knowledge graph: "acquired" belongs to film and television knowledge map, weather knowledge map and date knowledge map; "cloudy day" belongs to music knowledge map and weather knowledge map. And further determining the distance between the two service keywords in each knowledge graph to find that the distance between the two keywords in the weather knowledge graph is the minimum, so that the service field corresponding to the interactive statement is determined as weather query.
Fig. 7 is a schematic structural diagram of a terminal according to a first embodiment of the present invention. The terminal provided in this embodiment can implement the steps of the method applied to the terminal provided in any embodiment of the present invention. Specifically, the terminal provided in this embodiment includes:
the receiving module 11 is configured to receive an interactive statement input by a user, and acquire a keyword in the interactive statement;
a knowledge graph determining module 12, configured to determine a knowledge graph to which the keyword belongs, where the knowledge graph indicates a service domain of the keyword, the number of the keyword is at least two, and the number of the knowledge graph is at least two;
a distance determining module 13, configured to determine, for each of the at least two knowledge graphs, a distance of the interactive sentence within the knowledge graph, where the distance is a sum of distances of keywords of the interactive sentence within the knowledge graph;
a minimum value determining module 14, configured to determine a minimum value from the distances;
and the service field determining module 15 is configured to determine the service field corresponding to the interactive statement according to the distance.
The terminal provided by the embodiment of the invention extracts the key words from the interactive sentences input by the user; determining a knowledge graph to which the keyword belongs, then determining the distance of each keyword in the interactive sentence in each knowledge graph of at least two knowledge graphs, then determining the minimum value from the distances, and finally determining the business field corresponding to the interactive sentence according to the knowledge graph corresponding to the minimum value. In the process, the distance of the keywords in the knowledge graph indicates the distance of each keyword relation, namely the intimacy degree, so that the service field of the user tendency can be determined according to the minimum distance, the purpose of improving the positioning accuracy of the service field is realized, multiple rules and weights are written for the labels in the service field positioning process, and the complexity of the service field positioning is reduced.
Optionally, in an embodiment of the present invention, the distance determining module 13 is specifically configured to:
if all the keywords belong to one knowledge graph of the at least two knowledge graphs, determining the sum of the distances of the keywords in the knowledge graphs according to the connection relation between the keywords in the knowledge graphs, wherein the distances are finite values;
and determining the distance of the interactive sentences in the knowledge graph according to the sum of the distances of the keywords in the knowledge graph.
Further, optionally, the connection relationship includes a direct connection relationship and an indirect connection relationship, and a distance corresponding to the direct connection relationship is smaller than a distance corresponding to the indirect connection relationship.
Optionally, in an embodiment of the present invention, the distance determining module 13 is further specifically configured to:
if the keywords respectively belong to different knowledge graphs in the at least two knowledge graphs, determining that the distance between the keywords belonging to the knowledge graphs and the keywords belonging to other knowledge graphs is infinite for each knowledge graph in the at least two knowledge graphs;
and determining that the distance of the interactive statement in the knowledge graph is infinite, wherein the distance is infinite.
Fig. 8 is a schematic structural diagram of a second terminal embodiment of the present invention, and based on the above fig. 7, the terminal provided in this embodiment further includes:
a priority determining module 16, configured to determine, if distances of the interactive statements in the knowledge maps are the same, a knowledge map with a highest service priority from the at least two knowledge maps;
the service domain determining module 15 is specifically configured to determine the service domain corresponding to the interactive statement according to the knowledge graph with the highest service priority.
It should be noted that, the terminal according to the embodiment of the present invention is configured to execute the service domain positioning method, and the specific execution flow and beneficial effects are the same as those of the service domain positioning method, and the description of the present invention is omitted here.
Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (8)
1. A service domain positioning method is characterized by comprising the following steps:
receiving an interactive statement input by a user, and acquiring a keyword and a non-keyword in the interactive statement;
determining a knowledge graph to which the keyword belongs, wherein the knowledge graph indicates the service field of the keyword, the number of the keyword is at least two, and the number of the knowledge graph is at least two;
for each of the at least two knowledge graphs, determining a distance of the interactive sentence in the knowledge graph, wherein the distance is the sum of distances of keywords of the interactive sentence in the knowledge graph;
determining a minimum value from the distances;
determining the service field corresponding to the interactive statement according to the knowledge graph corresponding to the minimum value;
reducing the business in the business field corresponding to the interactive statement according to the non-keyword;
the determining the distance of the interactive statement within the knowledge-graph further comprises:
if the keywords respectively belong to different knowledge graphs in the at least two knowledge graphs, determining that the distance between the keywords belonging to the knowledge graphs and the keywords belonging to other knowledge graphs is infinite for each knowledge graph in the at least two knowledge graphs;
and determining that the distance of the interactive statement in the knowledge graph is infinite, wherein the distance is infinite.
2. The method of claim 1, wherein the determining the distance of the interactive statement within the knowledge-graph comprises:
if all the keywords belong to one knowledge graph of the at least two knowledge graphs, determining the sum of the distances of the keywords in the knowledge graphs according to the connection relation between the keywords in the knowledge graphs, wherein the distances are finite values;
and determining the distance of the interactive sentences in the knowledge graph according to the sum of the distances of the keywords in the knowledge graph.
3. The method according to claim 2, wherein the connection relationship comprises a direct connection relationship and an indirect connection relationship, and a distance corresponding to the direct connection relationship is smaller than a distance corresponding to the indirect connection relationship.
4. The method according to any one of claims 1 to 3, further comprising:
if the distance of the interactive statement in each knowledge graph is the same, determining the knowledge graph with the highest service priority from the at least two knowledge graphs;
and determining the service field corresponding to the interactive statement according to the knowledge graph with the highest service priority.
5. A service domain location terminal, comprising:
the receiving module is used for receiving interactive sentences input by users and acquiring keywords and non-keywords in the interactive sentences;
the knowledge graph determining module is used for determining a knowledge graph to which the keyword belongs, wherein the knowledge graph indicates the service field of the keyword, the number of the keyword is at least two, and the number of the knowledge graph is at least two;
a distance determining module, configured to determine, for each of the at least two knowledge graphs, a distance of the interactive sentence within the knowledge graph, where the distance is a sum of distances of keywords of the interactive sentence within the knowledge graph;
a minimum value determining module for determining a minimum value from the distances;
a service domain determining module, configured to determine, according to the knowledge graph corresponding to the minimum value, a service domain corresponding to the interactive statement;
the business field determining module is also used for reducing the business in the business field corresponding to the interactive statement according to the non-keyword;
the distance determining module is specifically further configured to:
if the keywords respectively belong to different knowledge graphs in the at least two knowledge graphs, determining that the distance between the keywords belonging to the knowledge graphs and the keywords belonging to other knowledge graphs is infinite for each knowledge graph in the at least two knowledge graphs;
and determining that the distance of the interactive statement in the knowledge graph is infinite, wherein the distance is infinite.
6. The terminal according to claim 5, wherein the distance determining module is specifically configured to:
if all the keywords belong to one knowledge graph of the at least two knowledge graphs, determining the sum of the distances of the keywords in the knowledge graphs according to the connection relation between the keywords in the knowledge graphs, wherein the distances are finite values;
and determining the distance of the interactive sentences in the knowledge graph according to the sum of the distances of the keywords in the knowledge graph.
7. The terminal according to claim 6, wherein the connection relationship comprises a direct connection relationship and an indirect connection relationship, and a distance corresponding to the direct connection relationship is smaller than a distance corresponding to the indirect connection relationship.
8. A terminal according to any of claims 5 to 7, further comprising:
the priority determining module is used for determining a knowledge graph with the highest service priority from the at least two knowledge graphs if the distances of the interactive statements in the knowledge graphs are the same;
the service domain determining module is specifically configured to determine a service domain corresponding to the interactive statement according to the knowledge graph with the highest service priority.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610545704.6A CN106126503B (en) | 2016-07-12 | 2016-07-12 | Service field positioning method and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610545704.6A CN106126503B (en) | 2016-07-12 | 2016-07-12 | Service field positioning method and terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106126503A CN106126503A (en) | 2016-11-16 |
| CN106126503B true CN106126503B (en) | 2020-02-11 |
Family
ID=57282496
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610545704.6A Active CN106126503B (en) | 2016-07-12 | 2016-07-12 | Service field positioning method and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106126503B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106354762B (en) * | 2016-08-17 | 2020-03-20 | 海信集团有限公司 | Service positioning method and device for interactive statements |
| CN106713083B (en) * | 2016-11-24 | 2020-06-26 | 海信集团有限公司 | Intelligent household equipment control method, device and system based on knowledge graph |
| CN110019824A (en) * | 2017-07-20 | 2019-07-16 | 苏州狗尾草智能科技有限公司 | Man-machine interaction method, the apparatus and system of knowledge based map |
| CN111177405A (en) * | 2019-12-18 | 2020-05-19 | 深圳壹账通智能科技有限公司 | Data search matching method and device, computer equipment and storage medium |
| CN111414764A (en) * | 2020-03-18 | 2020-07-14 | 苏州思必驰信息科技有限公司 | Method and system for determining skill field of dialog text |
| CN112749284B (en) * | 2020-12-31 | 2021-12-17 | 平安科技（深圳）有限公司 | Knowledge graph construction method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103324689A (en) * | 2013-06-04 | 2013-09-25 | 北京大学 | Business domain knowledge base building method and device |
| CN103425741A (en) * | 2013-07-16 | 2013-12-04 | 北京中科汇联信息技术有限公司 | Information exhibiting method and device |
| CN103995847A (en) * | 2014-05-06 | 2014-08-20 | 百度在线网络技术（北京）有限公司 | Information search method and device |
| CN104090955A (en) * | 2014-07-07 | 2014-10-08 | 科大讯飞股份有限公司 | Automatic audio/video label labeling method and system |
| CN105068661A (en) * | 2015-09-07 | 2015-11-18 | 百度在线网络技术（北京）有限公司 | Man-machine interaction method and system based on artificial intelligence |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8185558B1 (en) * | 2010-04-19 | 2012-05-22 | Facebook, Inc. | Automatically generating nodes and edges in an integrated social graph |
| KR101173561B1 (en) * | 2010-10-25 | 2012-08-13 | 한국전자통신연구원 | Question type and domain identifying apparatus and method |
| CN102314314B (en) * | 2011-08-29 | 2015-08-26 | 上海量明科技发展有限公司 | The method and system of cursor when in document reading conversion |
| US8484208B1 (en) * | 2012-02-16 | 2013-07-09 | Oracle International Corporation | Displaying results of keyword search over enterprise data |
-
2016
- 2016-07-12 CN CN201610545704.6A patent/CN106126503B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103324689A (en) * | 2013-06-04 | 2013-09-25 | 北京大学 | Business domain knowledge base building method and device |
| CN103425741A (en) * | 2013-07-16 | 2013-12-04 | 北京中科汇联信息技术有限公司 | Information exhibiting method and device |
| CN103995847A (en) * | 2014-05-06 | 2014-08-20 | 百度在线网络技术（北京）有限公司 | Information search method and device |
| CN104090955A (en) * | 2014-07-07 | 2014-10-08 | 科大讯飞股份有限公司 | Automatic audio/video label labeling method and system |
| CN105068661A (en) * | 2015-09-07 | 2015-11-18 | 百度在线网络技术（北京）有限公司 | Man-machine interaction method and system based on artificial intelligence |
Non-Patent Citations (3)
| Title |
|---|
| A subgraph matching tool for biological graphs;Yuanyuan Tian 等;《Bioinformatics》;20071231;第23卷(第2期);第232-239页 * |
| 一种面向多源知识图谱的样例查询方法;汤楠 等;《计算机研究与发展》;20151231(第S1期);参见第4页第4.1部分 * |
| 基于标记的图数据关键词查询方法研究;刘莹 等;《科学技术与工程》;20140930;第14卷(第26期);参见第259页第1段，第260页第3部分 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106126503A (en) | 2016-11-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106126503B (en) | Service field positioning method and terminal | |
| CN108920716B (en) | Data retrieval and visualization system and method based on knowledge graph | |
| CN107038207B (en) | Data query method, data processing method and device | |
| CN109213844B (en) | Text processing method and device and related equipment | |
| US10296160B2 (en) | Method for extracting salient dialog usage from live data | |
| CN109165302B (en) | Multimedia file recommendation method and device | |
| US10783885B2 (en) | Image display device, method for driving the same, and computer readable recording medium | |
| WO2021174783A1 (en) | Near-synonym pushing method and apparatus, electronic device, and medium | |
| CN110659366A (en) | Semantic analysis method and device, electronic equipment and storage medium | |
| US20120323948A1 (en) | Dialog-enhanced contextual search query analysis | |
| Ma et al. | Knowledge graph inference for spoken dialog systems | |
| CN110390094B (en) | Method, electronic device and computer program product for classifying documents | |
| CN107992608B (en) | SPARQL query statement automatic generation method based on keyword context | |
| CN110969022B (en) | Semantic determining method and related equipment | |
| CN112528001B (en) | Information query method and device and electronic equipment | |
| Sasidhar et al. | A survey on named entity recognition in Indian languages with particular reference to Telugu | |
| CN108875743B (en) | Text recognition method and device | |
| CN111309872B (en) | Search processing method, device and equipment | |
| KR102315181B1 (en) | Method, apparauts and system for named entity linking and computer program thereof | |
| CN107861948B (en) | Label extraction method, device, equipment and medium | |
| CN103942328A (en) | Video retrieval method and video device | |
| KR101369931B1 (en) | System and method for hybrid semantic searching service | |
| CN113343095A (en) | Model training and information recommendation method and device | |
| KR102345401B1 (en) | methods and apparatuses for content retrieval, devices and storage media | |
| CN105430536B (en) | A kind of video pushing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |