CN114780755A - Playing data positioning method and device based on knowledge graph and electronic equipment - Google Patents
Playing data positioning method and device based on knowledge graph and electronic equipment Download PDFInfo
- Publication number
- CN114780755A CN114780755A CN202210621599.5A CN202210621599A CN114780755A CN 114780755 A CN114780755 A CN 114780755A CN 202210621599 A CN202210621599 A CN 202210621599A CN 114780755 A CN114780755 A CN 114780755A
- Authority
- CN
- China
- Prior art keywords
- search
- keyword
- playing data
- preset
- data
- 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.)
- Pending
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
- 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
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- 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)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides a playing data positioning method and device based on a knowledge graph and electronic equipment. The play data positioning method comprises the following steps: acquiring at least one first search keyword to be positioned; inputting each first search keyword into a constructed knowledge map database, and determining a plurality of candidate search playing data matched with the first search keyword and a second search keyword which is semantically related to the first search keyword in each candidate search playing data; and screening target search playing data from the plurality of candidate search playing data based on the weight coefficient corresponding to each second search keyword, and determining the target moment of the second search keyword appearing in the target search playing data. According to the method and the device, the required target playing data can be directly found from numerous playing data and the desired playing position can be located, and the searching of the target playing data and the locating efficiency and accuracy of the target time in the target playing data are improved.
Description
Technical Field
The present application relates to the field of data processing technologies, and in particular, to a method and an apparatus for positioning playing data based on a knowledge graph, and an electronic device.
Background
With the development of electronic products, people have an increasing demand for data storage of electronic products, and therefore, the storage space of data in electronic products is increasing, but with the increasing amount of playing data required to be stored in electronic products, playing data such as audio data and video data is difficult to directly find required target playing data from a plurality of stored playing data and locate a desired playing position, and generally, a rough position can only be manually determined according to related memory of an operator, which results in a long time consumption and a low efficiency and accuracy for obtaining the target playing position of the target playing data.
Disclosure of Invention
In view of this, an object of the present application is to provide a playing data positioning method and apparatus based on a knowledge graph, and an electronic device, which can directly find needed target playing data from a plurality of playing data and position the target playing data to a desired playing position, thereby improving efficiency and accuracy of searching the target playing data and positioning a target time in the target playing data.
The embodiment of the application provides a playing data positioning method based on a knowledge graph, which comprises the following steps:
acquiring search information to be positioned;
performing word segmentation processing on the search information, and determining at least one first search keyword in the search information;
inputting each first search keyword into a constructed knowledge graph database, and determining a plurality of candidate search playing data matched with the first search keyword and a second search keyword which is semantically related to the first search keyword in each candidate search playing data; the knowledge graph database comprises a playing data knowledge graph, a weight coefficient of each preset keyword in each preset searching playing data and a preset moment when each preset keyword appears in each preset searching playing data;
and screening target search playing data from the plurality of candidate search playing data based on the weight coefficient corresponding to each second search keyword, and determining the target moment of the second search keyword appearing in the target search playing data.
Further, determining a weight coefficient of each preset keyword in each preset search playing data in the following manner;
inputting each preset search playing data into a feature extraction layer in a trained keyword scoring model, and determining at least one preset keyword in the preset search playing data;
and layering the semantic similarity of each preset keyword input into the trained keyword scoring model, and determining the weight coefficient of each preset keyword in the preset search playing data.
Further, the trained keyword scoring model is determined by the following method:
acquiring a plurality of sample searching and playing data and a sample label of each sample searching and playing data; the sample label is used for characterizing a real sample weight coefficient of each sample keyword in the corresponding sample search playing data in each sample search playing data;
inputting the sample search playing data and the sample label into an initial keyword scoring model, training the initial keyword scoring model, and determining a preset sample weight coefficient of each sample keyword in each sample search playing data;
and when the loss value between the preset sample weight coefficient and the real sample weight coefficient is smaller than a preset threshold value, stopping training and determining a trained keyword scoring model.
Further, the inputting each first search keyword into a constructed knowledge graph database, and determining a plurality of candidate search playing data matched with the first search keyword, includes:
and determining a plurality of candidate search playing data which are semantically related to the first search keyword or have a word frequency coupling degree larger than a preset value based on the first search keyword and a playing data knowledge graph in the knowledge graph database.
Further, the screening target search play data from the multiple candidate search play data based on the weight coefficient corresponding to each second search keyword includes:
screening out target search keywords of which the corresponding weight coefficients exceed preset weight coefficients from the plurality of second search keywords;
and determining the candidate search playing data corresponding to the target search keyword as the target search playing data.
The embodiment of the present application further provides a play data positioning device based on the knowledge graph, the play data positioning device includes:
the acquisition module is used for acquiring search information to be positioned;
the first determining module is used for performing word segmentation processing on the search information and determining at least one first search keyword in the search information;
the second determining module is used for inputting each first search keyword into a constructed knowledge graph database, and determining a plurality of candidate search playing data matched with the first search keyword and a second search keyword which is semantically related to the first search keyword in each candidate search playing data; the knowledge map database comprises a playing data knowledge map, a weight coefficient of each preset keyword in each preset searching playing data and a preset moment when each preset keyword appears in each preset searching playing data;
and the third determining module is used for screening target search playing data from the multiple candidate search playing data based on the weight coefficient corresponding to each second search keyword, and determining the target time of the second search keyword appearing in the target search playing data.
Further, the second determining module determines a weight coefficient of each preset keyword in each preset search play data in the following manner;
inputting each preset search playing data into a feature extraction layer in a trained keyword scoring model, and determining at least one preset keyword in the preset search playing data;
and inputting each preset keyword into a semantic similarity scoring layer in a trained keyword scoring model, and determining a weight coefficient of each preset keyword in preset search playing data.
Further, the trained keyword scoring model is determined by the following method:
obtaining a plurality of sample searching and playing data and a sample label of each sample searching and playing data; the sample label is used for representing a real sample weight coefficient of each sample keyword in the corresponding sample search playing data in each sample search playing data;
inputting the sample search playing data and the sample label into an initial keyword scoring model, training the initial keyword scoring model, and determining a preset sample weight coefficient of each sample keyword in each sample search playing data;
and when the loss value between the preset sample weight coefficient and the real sample weight coefficient is smaller than a preset threshold value, stopping training and determining a trained keyword scoring model.
An embodiment of the present application further provides an electronic device, including: the system comprises a processor, a memory and a bus, wherein the memory stores machine readable instructions executable by the processor, the processor and the memory are communicated through the bus when the electronic device runs, and the machine readable instructions are executed by the processor to execute the steps of the play data positioning method.
An embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the play data positioning method as described above are performed.
The playing data positioning method and device based on the knowledge graph and the electronic equipment provided by the embodiment of the application, compared with the prior art, the embodiment provided by the application determines a plurality of candidate search playing data matched with the first search keyword by inputting the first search keyword in the search information into the constructed knowledge map database, further determining target search play data from the plurality of candidate search play data according to a weight coefficient of a second search keyword semantically associated with the first search keyword, and determining a target time at which the second search keyword appears in the target search play data, the method and the device can directly find the needed target playing data from a plurality of playing data and locate the desired playing position, thereby improving the searching of the target playing data and the efficiency and the accuracy of locating the target moment in the target playing data.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
To more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.
Fig. 1 is a flowchart illustrating a method for locating play data based on a knowledge-graph according to an embodiment of the present application;
fig. 2 illustrates a second flowchart of a method for locating play data based on a knowledge-graph according to an embodiment of the present application;
FIG. 3 is a flowchart of an embodiment of a method for locating playback data based on a knowledge-graph according to an embodiment of the present application;
FIG. 4 is a schematic diagram illustrating a structure of a playback data positioning apparatus based on a knowledge-graph according to an embodiment of the present application;
fig. 5 shows a schematic structural diagram of an electronic device according to an embodiment of the present application.
In the figure:
400-play data positioning means; 410-an acquisition module; 420-a first determination module; 430-a second determination module; 440-a third determination module; 500-an electronic device; 510-a processor; 520-a memory; 530-bus.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, as presented in the figures, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. Every other embodiment that one skilled in the art can obtain without inventive effort based on the embodiments of the present application falls within the scope of protection of the present application.
First, an application scenario to which the present application is applicable will be described. Research shows that, in the prior art, with the increasing amount of playing data required to be stored in an electronic product, it is difficult for playing data such as audio data and video data to directly find required target playing data from a plurality of stored playing data and locate the desired playing position, and it is usually difficult to manually determine a rough position only according to the related memory of an operator, which results in a long time consumption and a low efficiency and accuracy for obtaining the target playing position of the target playing data.
Based on this, the embodiments of the present application provide a playing data positioning method and apparatus based on a knowledge graph, and an electronic device, which can directly find needed target playing data from a plurality of playing data and position the target playing data to a desired playing position, thereby improving efficiency and accuracy of target playing data search and positioning target time in the target playing data.
Referring to fig. 1, fig. 1 is a flowchart of a method for positioning playing data based on a knowledge graph according to an embodiment of the present application. As shown in fig. 1, the method for positioning playing data based on a knowledge graph provided in the embodiment of the present application includes the following steps:
s101, obtaining search information to be positioned.
In this step, the search information to be located may be characterized as a search statement that includes at least one first search keyword, and the search statement is text information under a general condition, and may also be audio information or video information if in a special application scenario.
Here, when the search term is audio information or video information, the audio information or the video information is subjected to character conversion by a voice recognition technology, converted into corresponding text information, and the converted text information is determined as search information to be positioned.
The special application scenarios include, but are not limited to, recording devices, video files, and persons who are inconvenient to perform related operations as entities corresponding to the entry of the search information.
Therefore, the search information to be positioned can be input by related workers, or can be input by unrelated workers who do not know the historical data corresponding to the search information at all, namely, a user of the play data positioning method provided by the application can not only be the related worker who can know the operation background, but also can be other workers who have the requirement of positioning the play data, and the limitation that the related play data positioning can only be carried out by the related workers who know the historical data in the traditional searching mode is avoided.
S102, performing word segmentation processing on the search information, and determining at least one first search keyword in the search information.
In the step, the search information is subjected to corresponding word segmentation processing and normalization processing according to conditions such as semantics and word frequency corresponding to sentences of the text information, so that at least one first search keyword in the search information is determined.
Here, the word segmentation of the search information according to the semantics corresponding to the sentence of the text information may specifically be:
and acquiring semantic information corresponding to the sentence of the text information where the search information is located and a preset standard vocabulary library.
And performing semantic word segmentation on the search information according to a preset standard vocabulary library, a blank space, the semantic information and punctuation marks, and determining at least one first search keyword in the search information.
Or, the word segmentation of the search information according to the word frequency corresponding to the sentence of the text information may specifically be:
and acquiring a preset standard vocabulary library.
And performing word segmentation on the search information according to a preset standard word library to determine words in the search information.
And sequencing and dividing the vocabulary in the search information according to the word frequency, and determining at least one first search keyword in the search information.
S103, inputting each first search keyword into a constructed knowledge map database, and determining a plurality of candidate search playing data matched with the first search keyword and a second search keyword which is semantically related to the first search keyword in each candidate search playing data; the knowledge graph database comprises a playing data knowledge graph, a weight coefficient of each preset keyword in each preset searching playing data and a preset moment when each preset keyword appears in each preset searching playing data.
In this step, since the constructed knowledge graph database not only stores the playing data knowledge graph, but also stores the weight coefficient of each preset keyword in each preset search playing data and the preset time when each preset keyword appears in each preset search playing data, after each first search keyword is input into the constructed knowledge graph database, each first search keyword firstly enters the playing data knowledge graph, a plurality of candidate search playing data matched with one search keyword are determined through the playing data knowledge graph, and then the weight coefficient of each preset keyword in each preset search playing data and the preset time when each preset keyword appears in each preset search playing data are determined through the trained keyword scoring model.
Here, the preset time at which the respective preset keywords appear in each preset search play data may be set as a label of the respective preset keywords.
The playing data knowledge graph in the knowledge graph database is constructed by performing data mining, information processing, knowledge measurement, semantic analysis, feature extraction and the like on single or multiple preset search playing data, and processing and analyzing entities, relations, attributes, relation attributes and the like to construct the playing data knowledge graph, and specifically comprises the following steps:
and carrying out semantic analysis on the single or multiple preset search playing data, and determining knowledge elements for constructing a playing data knowledge graph, wherein the knowledge elements include but are not limited to distribution of preset key words in the single or multiple preset search playing data, relations among the high-frequency words and attribute analysis among the preset key words.
And carrying out knowledge fusion on the knowledge elements, so that preset search playing data from different playing sources are subjected to data integration, disambiguation, processing, reasoning verification, updating and the like under the same frame specification, so that the mutual fusion of the knowledge elements and the frame specification to achieve the data, information, methods and experiences, the relevance among meanings is realized, and the constructed playing data knowledge map is determined.
In the foregoing, the embodiment that this application provided finds out more accurate candidate search broadcast data for the operating personnel through the broadcast data knowledge-graph that constructs, make more comprehensive data screening and provide the relevant information of more accurate broadcast data, compare in prior art, compare the broadcast data conversion throughout information storage such as characters and time, the embodiment that this application provided only needs the preset information of the preset keyword that the broadcast data knowledge-graph that the storage constructs abstracted, when having saved a large amount of memory resources, still provided the operating efficiency, guaranteed the accuracy of operation.
Further, in step S103, the weight coefficient of each preset keyword in each preset search play data is determined through the following sub-steps:
and a substep 1031 of inputting each preset search playing data into a feature extraction layer in the trained keyword scoring model, and determining at least one preset keyword in the preset search playing data.
Here, first, the data feature of each preset search playing data is extracted, and at least one preset keyword in each preset search playing data is determined.
For example, in the embodiment provided by the present application, the feature extraction of the feature extraction layer may be performed according to, but not limited to, semantics and word frequency of each preset search play data.
And a substep 1032 of inputting each preset keyword into a semantic similarity grading layer in the trained keyword grading model, and determining a weight coefficient of each preset keyword in the preset search playing data.
Here, the trained keyword scoring model is determined by:
obtaining a plurality of sample searching and playing data and a sample label of each sample searching and playing data; the sample label is used for characterizing a real sample weight coefficient of each sample keyword in the corresponding sample search playing data in each sample search playing data.
Inputting the sample searching and playing data and the sample label into an initial keyword scoring model, training the initial keyword scoring model, and determining a preset sample weight coefficient of each sample keyword in each sample searching and playing data.
And when the loss value between the preset sample weight coefficient and the real sample weight coefficient is smaller than a preset threshold value, stopping training and determining a trained keyword scoring model.
In the above, after determining the weight coefficient of each preset keyword in the preset search playing data, the weight coefficient of each preset keyword in each preset search playing data and the preset time when each preset keyword appears in each preset search playing data are all stored in the knowledge map database.
Here, the preset time is time information in which each preset keyword appears in the corresponding preset search play data.
Further, in step S103, inputting each first search keyword into a knowledge graph database constructed, and determining a plurality of candidate search playing data matched with the first search keyword, including:
and determining a plurality of candidate search playing data which are semantically related to the first search keyword or have a word frequency coupling degree larger than a preset value based on the first search keyword and a playing data knowledge graph in the knowledge graph database.
Before determining a plurality of candidate search playing data, firstly setting a preset coupling degree value, at the moment, inputting a playing data knowledge graph in a knowledge graph database by a first search keyword, searching for the search playing data with semantic association or word frequency coupling degree with the first search keyword, then determining a real semantic association value and a word frequency coupling value between the search playing data and the first search keyword by using a weighted average mode of the semantic association or the word frequency coupling degree or other self-defined calculation modes, and determining the search playing data larger than the preset value as the plurality of candidate search playing data matched with the first search keyword when the real semantic association value and the word frequency coupling value are larger than the preset value.
The candidate search playing data may be one or more, and the word frequency coupling value between the search playing data and the first search keyword is determined based on the word repetition degree between the search playing data and the first search keyword, specifically:
and determining the number of words semantically associated with the first search keyword in the search playing data.
And determining the ratio of the number of the words to the first search keyword as a word frequency coupling value between the search playing data and the first search keyword.
In the above, the plurality of candidate search broadcast data aggregations semantically associated with the first search keyword may quantify semantic relation between semantic nodes through semantic symbiosis.
The higher the semantic symbiosis between the two semantic nodes is, the higher the probability that the two first search keywords appear in the same playing data is.
S104, screening target search playing data from the candidate search playing data based on the weight coefficient corresponding to each second search keyword, and determining the target time of the second search keyword appearing in the target search playing data.
In the step, the second search keyword is a search keyword in the preset keywords having semantic association with the first search keyword, wherein the weight coefficient book of the second search keyword is stored in the constructed knowledge graph database, after a plurality of candidate search play data are determined, the plurality of candidate search play data are sequentially displayed according to the weight coefficient corresponding to the second search keyword, a worker can directly check the weight coefficient corresponding to each second search keyword in a candidate list consisting of the plurality of candidate search play data, at this time, the worker can select the second search keyword of a certain or any one of the candidate search play data according to actual requirements, at this time, the selected candidate search play data is the target search play data, and the weight coefficient corresponding to the second search keyword in the target search play data is certainly greater than the preset weight coefficient, at the moment, the worker can directly operate the second search keyword through any operation instruction, and at the moment, the target search playing data can be directly positioned at the target moment when the second search keyword appears.
Here, after determining a target time at which the second search keyword appears in the target search play data, displaying the target search play data related to the second search keyword staying at the target time to the staff, where the content displayed to the staff is described in the following specific embodiment: if the audio and video of the conference record which is semantically related to the second search keyword in the preset search playing data can be displayed; or, the educational learning audio and video data which are semantically related to the second search keyword in the preset search playing data can be displayed.
Compared with the prior art, the playing data positioning method provided by the embodiment of the application has the advantages that the first search keyword in the search information is input into the constructed knowledge graph database, the plurality of candidate search playing data matched with the first search keyword are determined, the target search playing data are further determined from the plurality of candidate search playing data according to the weight coefficient of the second search keyword in semantic association with the first search keyword, the target moment of the second search keyword appearing in the target search playing data is determined, the required target playing data can be directly found from the plurality of playing data and the desired playing position can be positioned, and the target playing data search efficiency and the target moment positioning accuracy in the target playing data are improved.
Referring to fig. 2, fig. 2 is a second flowchart of a method for positioning playing data based on a knowledge graph according to the present application. As shown in fig. 2, a method for positioning play data based on a knowledge graph according to an embodiment of the present application includes:
s201, obtaining search information to be positioned.
S202, performing word segmentation processing on the search information, and determining at least one first search keyword in the search information.
S203, inputting each first search keyword into a constructed knowledge map database, and determining a plurality of candidate search playing data matched with the first search keyword and a second search keyword which is semantically related to the first search keyword in each candidate search playing data; the knowledge graph database comprises a playing data knowledge graph, a weight coefficient of each preset keyword in each preset searching playing data and a preset moment when each preset keyword appears in each preset searching playing data.
S204, screening out target search keywords of which the corresponding weight coefficients exceed preset weight coefficients from the plurality of second search keywords.
In the step, at least one second search keyword with the weight coefficient exceeding the preset weight coefficient is taken as a target search keyword.
Here, the preset weighting factor may be set by user according to actual needs of different application scenarios.
The preset weight coefficient processing of the second search keyword is set according to the conventional semantic association degree and the word frequency coupling degree, and the weighted weight coefficient setting can be carried out by combining the time duration and the time position information of the search playing data corresponding to the second search keyword, so that the weight coefficient of the second search keyword is finally determined.
S205, determining the candidate search playing data corresponding to the target search keyword as the target search playing data, and determining the target time of the second search keyword appearing in the target search playing data.
Here, after the target search play data is determined, the operator may select the second search keyword according to the actual application scenario, and directly locate the target time appearing in the target search play data by selecting the second search keyword.
The descriptions of S201 to S203 may refer to the descriptions of S101 to S103, and the same technical effects can be achieved, which are not described in detail.
Here, a specific flow of the play data positioning method provided by the present application is described through an embodiment, as shown in fig. 3, fig. 3 is a flow chart of an embodiment of the play data positioning method based on a knowledge graph provided by the present application according to the present application, where the embodiment includes the following steps:
s301, extracting audio and video characteristics in preset searching and playing data, mining data, processing information and measuring knowledge, and determining initial knowledge map elements.
S302, inputting preset search playing data into a preset knowledge graph framework platform for semantic analysis, determining distribution of each preset keyword, relationship among high-frequency preset keywords in each preset keyword and attribute analysis among the preset keywords, updating an initial knowledge graph element according to the analysis result, and determining a target knowledge graph element.
And S303, inputting the target knowledge graph element into a preset knowledge graph framework platform for knowledge fusion, and determining the playing data knowledge graph.
The target knowledge graph elements are input into a preset knowledge graph framework platform to perform knowledge fusion, so that playing data from different sources are subjected to data integration, disambiguation, processing, reasoning verification and updating under the same framework specification, the playing data are fused with data, information, methods, experiences and ideas, and the playing data knowledge graph is formed by using semantic relevance among preset keywords.
S304, based on the trained keyword scoring model, determining a weight coefficient of each preset keyword in the playing data knowledge graph, and recording a preset moment when each preset keyword appears in each preset search playing data.
S305, storing the preset time when each preset keyword appears in each preset search playing data, the weight coefficient of each preset keyword and the playing data knowledge graph in a constructed knowledge graph database together.
S306, performing word segmentation and normalization processing on the to-be-positioned search information input by the staff, and determining at least one first search keyword in the search information.
S307, inputting each first search keyword into the constructed knowledge graph database, and determining a plurality of candidate search playing data matched with the first search keywords and second search keywords having semantic relevance with the first search keywords in each candidate search playing data.
S308, the second search keywords are ranked according to the weight coefficients, relevant candidate search play is displayed, target search play data are screened from the candidate search play data, target time of the second search keywords appearing in the target search play data is determined, and the target search play data are positioned to the target time.
Compared with the prior art, the playing data positioning method provided by the embodiment of the application has the advantages that the first search keyword in the search information is input into the constructed knowledge graph database, the plurality of candidate search playing data matched with the first search keyword are determined, the target search playing data are further determined from the plurality of candidate search playing data according to the weight coefficient of the second search keyword in semantic association with the first search keyword, the target moment of the second search keyword appearing in the target search playing data is determined, the required target playing data can be directly found from the plurality of playing data and the desired playing position can be positioned, and the target playing data search efficiency and the target moment positioning accuracy in the target playing data are improved.
Referring to fig. 4, fig. 4 is a schematic structural diagram of a playback data positioning device based on a knowledge graph according to an embodiment of the present application. As shown in fig. 4, the play data positioning apparatus 400 includes:
the obtaining module 410 is configured to obtain search information to be located.
The first determining module 420 is configured to perform word segmentation processing on the search information, and determine at least one first search keyword in the search information.
A second determining module 430, configured to input each first search keyword into a constructed knowledge graph database, and determine a plurality of candidate search playing data that match the first search keyword, and a second search keyword that is semantically associated with the first search keyword in each candidate search playing data; the knowledge graph database comprises a playing data knowledge graph, a weight coefficient of each preset keyword in each preset searching playing data and a preset moment when each preset keyword appears in each preset searching playing data.
The third determining module 440 is configured to screen target search playing data from the multiple candidate search playing data based on the weight coefficient corresponding to each second search keyword, and determine a target time when the second search keyword appears in the target search playing data.
Further, the second determining module 430 determines the weight coefficient of each preset keyword in each preset search playing data specifically in the following manner.
And inputting each preset search playing data into a feature extraction layer in a trained keyword scoring model, and determining at least one preset keyword in the preset search playing data.
And inputting each preset keyword into a semantic similarity scoring layer in a trained keyword scoring model, and determining a weight coefficient of each preset keyword in preset search playing data.
Further, the trained keyword scoring model is determined in the following manner.
Acquiring a plurality of sample searching and playing data and a sample label of each sample searching and playing data; the sample label is used for characterizing a real sample weight coefficient of each sample keyword in the corresponding sample search playing data in each sample search playing data.
Inputting the sample searching and playing data and the sample label into an initial keyword scoring model, training the initial keyword scoring model, and determining a preset sample weight coefficient of each sample keyword in each sample searching and playing data.
And when the loss value between the preset sample weight coefficient and the real sample weight coefficient is smaller than a preset threshold value, stopping training and determining a trained keyword scoring model.
Compared with the prior art, the play data positioning apparatus 400 provided by the embodiment of the present application determines a plurality of candidate search play data matching a first search keyword in search information by inputting the first search keyword into a constructed knowledge graph database, further determines target search play data from the plurality of candidate search play data according to a weight coefficient of a second search keyword having semantic association with the first search keyword, and determines a target time of the second search keyword appearing in the target search play data, and can directly find needed target play data from a plurality of play data and position the needed target play data to a desired play position, thereby improving efficiency and accuracy of target play data search and positioning of the target time in the target play data.
Referring to fig. 5, fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure. As shown in fig. 5, the electronic device 500 includes a processor 510, a memory 520, and a bus 530.
The memory 520 stores machine-readable instructions executable by the processor 510, when the electronic device 500 runs, the processor 510 communicates with the memory 520 through the bus 530, and when the machine-readable instructions are executed by the processor 510, the steps of the play data positioning method in the method embodiments shown in fig. 1 and fig. 2 may be executed.
An embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the step of the play data positioning method in the method embodiments shown in fig. 1 and fig. 2 may be executed.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed coupling or direct coupling or communication connection between each other may be through some communication interfaces, indirect coupling or communication connection between devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in software functional units and sold or used as a stand-alone product, may be stored in a non-transitory computer-readable storage medium executable by a processor. Based on such understanding, the technical solutions of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present application, and are used for illustrating the technical solutions of the present application, but not limiting the same, and the scope of the present application is not limited thereto, and although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope disclosed in the present application; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present application and are intended to be covered by the appended claims. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A playing data positioning method based on a knowledge graph is characterized by comprising the following steps:
acquiring search information to be positioned;
performing word segmentation processing on the search information, and determining at least one first search keyword in the search information;
inputting each first search keyword into a constructed knowledge graph database, and determining a plurality of candidate search playing data matched with the first search keyword and a second search keyword which is semantically related to the first search keyword in each candidate search playing data; the knowledge map database comprises a playing data knowledge map, a weight coefficient of each preset keyword in each preset searching playing data and a preset moment when each preset keyword appears in each preset searching playing data;
and screening target search playing data from the plurality of candidate search playing data based on the weight coefficient corresponding to each second search keyword, and determining the target moment of the second search keyword appearing in the target search playing data.
2. The playback data locating method according to claim 1, characterized in that the weighting factor of each preset keyword in each preset search playback data is determined by;
inputting each preset search playing data into a feature extraction layer in a trained keyword scoring model, and determining at least one preset keyword in the preset search playing data;
and layering the semantic similarity of each preset keyword input into the trained keyword scoring model, and determining the weight coefficient of each preset keyword in the preset search playing data.
3. The method of claim 2, wherein the trained keyword scoring model is determined by:
acquiring a plurality of sample searching and playing data and a sample label of each sample searching and playing data; the sample label is used for characterizing a real sample weight coefficient of each sample keyword in the corresponding sample search playing data in each sample search playing data;
inputting the sample search playing data and the sample label into an initial keyword scoring model, training the initial keyword scoring model, and determining a preset sample weight coefficient of each sample keyword in each sample search playing data;
and when the loss value between the preset sample weight coefficient and the real sample weight coefficient is smaller than a preset threshold value, stopping training and determining a trained keyword scoring model.
4. The method for locating playing data according to claim 1, wherein the inputting each first search keyword into a constructed knowledge map database, and determining a plurality of candidate search playing data matching the first search keyword comprises:
and determining a plurality of candidate search playing data which are semantically related to the first search keyword or have a word frequency coupling degree larger than a preset value based on the first search keyword and a playing data knowledge graph in the knowledge graph database.
5. The method for locating playing data according to claim 1, wherein the step of screening target search playing data from a plurality of candidate search playing data based on the weight coefficient corresponding to each second search keyword comprises:
screening out target search keywords of which the corresponding weight coefficients exceed preset weight coefficients from the plurality of second search keywords;
and determining the candidate search playing data corresponding to the target search keyword as the target search playing data.
6. A playback data positioning apparatus based on a knowledge-graph, the playback data positioning apparatus comprising:
the acquisition module is used for acquiring search information to be positioned;
the first determining module is used for performing word segmentation processing on the search information and determining at least one first search keyword in the search information;
the second determining module is used for inputting each first search keyword into a constructed knowledge map database, determining a plurality of candidate search playing data matched with the first search keyword and a second search keyword which is semantically related to the first search keyword in each candidate search playing data; the knowledge map database comprises a playing data knowledge map, a weight coefficient of each preset keyword in each preset searching playing data and a preset moment when each preset keyword appears in each preset searching playing data;
and the third determining module is used for screening target search playing data from the multiple candidate search playing data based on the weight coefficient corresponding to each second search keyword, and determining the target time of the second search keyword appearing in the target search playing data.
7. The playback data positioning apparatus according to claim 6, wherein the second determining module determines the weighting factor of each preset keyword in each preset search playback data in the following manner;
inputting each preset search playing data into a feature extraction layer in a trained keyword scoring model, and determining at least one preset keyword in the preset search playing data;
and inputting each preset keyword into a semantic similarity scoring layer in a trained keyword scoring model, and determining a weight coefficient of each preset keyword in preset search playing data.
8. The playback data positioning apparatus of claim 7, wherein the trained keyword scoring model is determined by:
obtaining a plurality of sample searching and playing data and a sample label of each sample searching and playing data; the sample label is used for representing a real sample weight coefficient of each sample keyword in the corresponding sample search playing data in each sample search playing data;
inputting the sample search playing data and the sample label into an initial keyword scoring model, training the initial keyword scoring model, and determining a preset sample weight coefficient of each sample keyword in each sample search playing data;
and when the loss value between the preset sample weight coefficient and the real sample weight coefficient is smaller than a preset threshold value, stopping training and determining a trained keyword scoring model.
9. An electronic device, comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, the processor and the memory communicate via the bus when the electronic device is running, and the machine-readable instructions are executed by the processor to perform the steps of the play data positioning method according to any one of claims 1 to 5.
10. A computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program performs the steps of the play data positioning method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210621599.5A CN114780755A (en) | 2022-06-01 | 2022-06-01 | Playing data positioning method and device based on knowledge graph and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210621599.5A CN114780755A (en) | 2022-06-01 | 2022-06-01 | Playing data positioning method and device based on knowledge graph and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114780755A true CN114780755A (en) | 2022-07-22 |
Family
ID=82420777
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210621599.5A Pending CN114780755A (en) | 2022-06-01 | 2022-06-01 | Playing data positioning method and device based on knowledge graph and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114780755A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114996482A (en) * | 2022-08-03 | 2022-09-02 | 北京达佳互联信息技术有限公司 | Knowledge graph construction method, knowledge graph construction device, video search method, device and electronic equipment |
-
2022
- 2022-06-01 CN CN202210621599.5A patent/CN114780755A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114996482A (en) * | 2022-08-03 | 2022-09-02 | 北京达佳互联信息技术有限公司 | Knowledge graph construction method, knowledge graph construction device, video search method, device and electronic equipment |
| CN114996482B (en) * | 2022-08-03 | 2022-11-11 | 北京达佳互联信息技术有限公司 | Knowledge graph construction method, knowledge graph construction device, video search method, device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106649786B (en) | Answer retrieval method and device based on deep question answering | |
| CN102119385B (en) | Method and subsystem for searching media content within a content-search-service system | |
| CN112667794A (en) | Intelligent question-answer matching method and system based on twin network BERT model | |
| US8606779B2 (en) | Search method, similarity calculation method, similarity calculation, same document matching system, and program thereof | |
| CN109783631B (en) | Community question-answer data verification method and device, computer equipment and storage medium | |
| CN111324771B (en) | Video tag determination method and device, electronic equipment and storage medium | |
| CN109558513B (en) | Content recommendation method, device, terminal and storage medium | |
| US11531693B2 (en) | Information processing apparatus, method and non-transitory computer readable medium | |
| CN111369980B (en) | Voice detection method, device, electronic equipment and storage medium | |
| CN111259262A (en) | Information retrieval method, device, equipment and medium | |
| CN111767334A (en) | Information extraction method and device, electronic equipment and storage medium | |
| CN107844531B (en) | Answer output method and device and computer equipment | |
| CN117112754A (en) | Information processing method, information processing device, electronic equipment and storage medium | |
| CN114780755A (en) | Playing data positioning method and device based on knowledge graph and electronic equipment | |
| JP2009199302A (en) | Program, device, and method for analyzing document | |
| CN116402166B (en) | Training method and device of prediction model, electronic equipment and storage medium | |
| CN117520800A (en) | Training method, system, electronic equipment and medium for nutrition literature model | |
| CN111881695A (en) | Audit knowledge retrieval method and device | |
| CN116342082A (en) | Knowledge graph-based post competence judging method, device, medium and equipment | |
| JP5877775B2 (en) | Content management apparatus, content management system, content management method, program, and storage medium | |
| JP2017027168A (en) | Taste learning method, taste learning program and taste learning device | |
| CN115048536A (en) | Knowledge graph generation method and device, computer equipment and storage medium | |
| CN115292506A (en) | Knowledge graph ontology construction method and device applied to office field | |
| CN114201607A (en) | Information processing method and device | |
| CN109284364B (en) | Interactive vocabulary updating method and device for voice microphone-connecting interaction |
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 |