A display device and a label template library generation method

Abstract

The present application provides a display device and a method for generating a label template library. The method obtains test data in response to a control instruction for building a label template library, performs word segmentation processing on the test data, and obtains a word segmentation result with a category label. Generate a label template according to the category label, and cluster the label template to obtain a cluster template set. Perform a test check on the cluster template set according to the test data, and filter out a first template set to generate a label template library. The present application performs word segmentation on the test data, generates a label template according to the word segmentation result, and clusters the label template to generate a label template library according to the cluster template set, thereby solving the problem of time-consuming manual writing and low efficiency in generating a label template library.

Description

Technical Field

[0001] The present application relates to the technical field of display devices, and in particular to a display device and a label template library generation method. Background Art

[0002] In order to improve the convenience of the display device, the display device is provided with a voice control function, that is, the display device can receive voice commands sent by the user and parse the voice commands, and then execute the voice interaction process of the display device according to the results of the analysis to realize voice control.

[0003] In the process of parsing the voice command, the display device can match the keywords in the voice command through the label template, and then parse the voice command into a machine command that can be executed by the display device according to the matching result, thereby completing the parsing process.

[0004] Label templates need to be written manually. For entries with a large amount of data, writing label templates takes a lot of time. Therefore, generating a label template library also takes a lot of time, and the efficiency of generating a label template library is low. Summary of the invention

[0005] The present application provides a display device and a label template library generation method to solve the problem of low efficiency in generating a label template library.

[0006] In a first aspect, some embodiments of the present application provide a display device, comprising: a display configured to display a user interface; a memory configured to store test data; and a controller configured to:

[0007] Responding to a control instruction for building a label template library, acquiring test data;

[0008] Performing word segmentation processing on the test data to obtain a word segmentation result, wherein the word segmentation result includes a word segmentation and a category label corresponding to the word segmentation;

[0009] Sorting the category labels to obtain a label template;

[0010] Performing clustering on the label template to obtain at least one cluster template set, wherein the cluster template set has corresponding cluster labels;

[0011] Matching the semantic information of the test data with the clustering labels corresponding to the clustering template set to obtain a matching probability between the clustering template set and the test data;

[0012] Filtering out a first template set whose matching probability is greater than or equal to a probability threshold from the at least one clustering template set;

[0013] The label template library is generated according to the first template set.

[0014] The above technical solution has the following advantages or beneficial effects: label templates with the same or similar structures are screened out through clustering for merging, and label templates with relatively high matching probability are screened out, thereby generating a label template library based on the label templates, thereby improving the generation efficiency of the label template library.

[0015] In some embodiments, the controller performs the step of clustering the label templates to obtain a cluster template set, which is specifically configured as follows:

[0016] Calculating first similarities between different label templates according to the category relationship between the corresponding positions of the category labels in the label templates;

[0017] If the first similarity is greater than or equal to a similarity threshold, selecting a target label template from label templates corresponding to the first similarity;

[0018] Generating a clustering label according to the template structure of the target label template;

[0019] Clustering and merging the label templates to obtain the clustering template set;

[0020] Add the corresponding clustering label to the clustering template set.

[0021] The above technical solution has the following advantages or beneficial effects: for different label templates, by calculating the similarity between label templates, label templates with relatively high similarity are clustered and merged. After merging, among the label templates that have completed clustering and merging, the category label is selected as the set label of the cluster template set. By clustering label templates, the operations on label templates of the same category are normalized, and the integration efficiency of label templates is improved.

[0022] In some embodiments, the controller clusters and merges the label templates to obtain the cluster template set, which is specifically configured as follows:

[0023] Calculating a second similarity according to the category relationship between the category labels at corresponding positions in the label template and the cluster template, wherein the cluster template is a label template that has completed cluster merging;

[0024] If the second similarity is greater than or equal to the similarity threshold, the label template is merged into the clustering template set.

[0025] The above technical solution has the following advantages or beneficial effects: after clustering and merging, the clustering template set is taken as a category as a whole, and the second similarity between the label template and the clustering template set is calculated to merge the label template again according to the second similarity.

[0026] In some embodiments, the controller performs the step of sorting the category labels to obtain a label template, which is specifically configured as follows:

[0027] Obtaining the word segmentation order of the word segmentation corresponding to each of the word segmentation results;

[0028] According to the word segmentation order, the category labels in the corresponding word segmentation results are sorted to obtain a sorting result;

[0029] A label template is generated according to the sorting result.

[0030] The above technical solution has the following advantages or beneficial effects: according to the word segmentation order of the word segmentation results, the category labels of different word segmentation results in the same test data are merged to generate a label template corresponding to the test data.

[0031] In some embodiments, the controller performs the step of matching the semantic information of the test data with the clustering labels corresponding to the clustering template set, which is specifically configured as follows:

[0032] Obtaining a first matching number of the label templates in the clustering template set whose semantic similarity with the semantic information of the test data is greater than or equal to a semantic similarity threshold;

[0033] Summing each of the first matching times to obtain the number of label matches where the clustering label matches the semantic information of the test data;

[0034] Obtaining the total number of matches of the label templates in each of the clustering template sets, whose semantic similarity with the semantic information of the test data is greater than or equal to a semantic similarity threshold;

[0035] The matching probability between the test data and the clustering template set is calculated according to the number of label matches and the total number of matches.

[0036] The above technical solution has the following advantages or beneficial effects: voice commands can express the same interactive meaning through different word orders. Therefore, by changing the order of label templates, the matching probability of the same label template under different category label orderings is calculated, thereby improving the accuracy of calculating the matching probability of the label template.

[0037] In some embodiments, the controller is further configured to:

[0038] Changing the order of the category labels in the label template to obtain a random label template;

[0039] Obtaining a second matching number of the out-of-order label template that matches the test data;

[0040] The first matching number is updated according to the second matching number.

[0041] In some embodiments, after the controller executes the step of changing the arrangement order of the category labels in the label template, the controller is further configured to:

[0042] Detecting a format tag of the label template, where the format tag is a category tag in the label template used to represent format information;

[0043] If no format tag is detected in the tag template, generating a format tag according to other category tags in the tag template;

[0044] Add the format tag in the tag template.

[0045] The above technical solution has the following advantages or beneficial effects: if the label template does not have a format tag, the display device cannot perform the control interaction process according to the label template. When there is no format tag in the label template, the format tag can be generated according to the label attributes of other category tags, and the format tag can be added to the label template.

[0046] In some embodiments, after the controller performs the step of matching the semantic information of the test data with the clustering labels corresponding to the clustering template set, the controller is further configured to:

[0047] Obtain invalid test data from each of the test data, wherein the invalid test data is test data whose semantic similarity between semantic information and the label template is less than a semantic similarity threshold;

[0048] Performing iterative word segmentation on the invalid test data to obtain an iterative word segmentation result, wherein the iterative word segmentation result includes an iterative word segmentation and an iterative category label corresponding to the iterative word segmentation;

[0049] Generate an iteration label template based on the iteration category label.

[0050] The above technical solution has the following advantages or beneficial effects: by performing iterative word segmentation and template generation steps on invalid test data that has not been successfully matched, the extensiveness of template generation is improved.

[0051] In some embodiments, the controller performs the step of clustering and merging the label templates, which is specifically configured as follows:

[0052] Obtaining the number of clusters, where the number of clusters is the maximum number of the clustering template set;

[0053] In the process of clustering and merging the label templates, obtaining the number of the cluster template sets;

[0054] When the number of clustering template sets is equal to the number of clusters, the label template clustering and merging into a new clustering template set is stopped.

[0055] The above technical solution has the following advantages or beneficial effects: by setting the number of clusters, the total number of cluster template sets is determined, and when the number of cluster template sets is equal to the total number, cluster merging of label templates is stopped to ensure the diversity of label template categories.

[0056] In a second aspect, some embodiments of the present application provide a method for generating a label template library, which is applied to the display device described in the first aspect, wherein the display device includes a display, a memory, and a controller, wherein the display is configured to display a user interface; the memory is configured to store test data; and the method includes:

[0057] Responding to a control instruction for building a label template library, acquiring test data;

[0058] Performing word segmentation processing on the test data to obtain a word segmentation result, wherein the word segmentation result includes a word segmentation and a category label corresponding to the word segmentation;

[0059] Sorting the category labels to obtain a label template;

[0060] Performing clustering on the label template to obtain at least one cluster template set, wherein the cluster template set has corresponding cluster labels;

[0061] Matching the semantic information of the test data with the clustering labels corresponding to the clustering template set to obtain a matching probability between the clustering template set and the test data;

[0062] Filtering out a first template set whose matching probability is greater than or equal to a probability threshold from the at least one clustering template set;

[0063] The label template library is generated according to the first template set.

[0064] It can be seen from the above technical solutions that the present application provides a display device and a label template library generation method, wherein the method obtains test data in response to a control instruction for building a label template library, performs word segmentation processing on the test data, and obtains a word segmentation result with a category label. A label template is generated according to the category label, and the label template is clustered to obtain a cluster template set. A test verification is performed on the cluster template set according to the test data, a first template set is screened out, and the label templates are clustered to generate a label template library according to the cluster template set, thereby solving the problem of time-consuming manual writing and low efficiency in generating a label template library. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0066] Figure 1 A schematic diagram of an operation scenario between a display device and a control device provided in some embodiments of the present application;

[0067] Figure 2 A schematic diagram of the hardware configuration of a display device provided in some embodiments of the present application;

[0068] Figure 3 A schematic diagram of software configuration of a display device provided in some embodiments of the present application;

[0069] Figure 4 A flowchart of a method for generating a label template library by a display device provided in some embodiments of the present application;

[0070] Figure 5 A timing diagram of a method for generating a label template library executed by a display device provided in some embodiments of the present application;

[0071] Figure 6 A schematic diagram of assigning category labels to display devices provided in some embodiments of the present application;

[0072] Figure 7 A flowchart of generating a label template for a display device provided in some embodiments of the present application;

[0073] Figure 8 A flow chart of assigning multiple labels to a word segmentation result by a display device provided in some embodiments of the present application;

[0074] Fig. 9 A flowchart of a first embodiment of a display device performing merging and clustering according to some embodiments of the present application;

[0075] Fig.10 A flowchart of a second embodiment of a display device performing merging and clustering according to some embodiments of the present application;

[0076] Fig.11 A flow chart of calculating the matching probability of a cluster template set by a display device provided in some embodiments of the present application;

[0077] Fig.12 A flow chart of classifying a cluster template set by a display device provided in some embodiments of the present application;

[0078] Fig.13 A flowchart of updating the first matching number according to the out-of-order label template in some embodiments of the present application;

[0079] Fig.14 A flowchart of limiting the number of sets according to the number of set clusters in some embodiments of the present application. DETAILED DESCRIPTION

[0080] The following embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following embodiments do not represent all implementations consistent with the present application. They are only examples of systems and methods consistent with some aspects of the present application as detailed in the claims.

[0081] It should be noted that the brief description of terms in this application is only for the convenience of understanding the embodiments described below, and is not intended to limit the embodiments of this application. Unless otherwise specified, these terms should be understood according to their ordinary and common meanings.

[0082] The terms "first", "second", "third", etc. in the specification and claims of this application and the above drawings are used to distinguish similar or similar objects or entities, and do not necessarily mean to limit a specific order or sequence, unless otherwise noted. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances.

[0083] The terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover but not exclude inclusion, for example, a product or device comprising a list of components is not necessarily limited to all the components expressly listed but may include other components not expressly listed or inherent to such product or device.

[0084] The term "module" refers to any known or later developed hardware, software, firmware, artificial intelligence, fuzzy logic, or combination of hardware and / or software code that is capable of performing the functions associated with that element.

[0085] In the embodiment of the present application, the display device 200 generally refers to a device with image display and data processing capabilities. For example, the display device 200 includes but is not limited to a smart TV, a mobile terminal, a computer, a monitor, an advertising screen, a wearable device, a virtual reality device, an augmented reality device, etc.

[0086] Figure 1 This is a schematic diagram of an operation scenario between a display device and a control device provided in some embodiments of the present application. Figure 1 As shown in FIG. 1 , a user can operate the display device 200 through touch operation, the mobile terminal 300 and the control device 100. The control device 100 is used to receive operation instructions input by the user and convert the operation instructions into control instructions that can be recognized and responded to by the display device 200. For example, the control device 100 can be a remote controller, a stylus pen, a handle, etc.

[0087] The mobile terminal 300 can be used as a control device for performing human-computer interaction between the user and the display device 200. The mobile terminal 300 can also be used as a communication device for establishing a communication connection with the display device 200 and performing data interaction. In some embodiments, the mobile terminal 300 can install software applications with the display device 200, and achieve connection and communication through a network communication protocol to achieve the purpose of one-to-one control operation and data communication. The audio and video content displayed on the mobile terminal 300 can also be transmitted to the display device 200 to achieve a synchronous display function.

[0088] In some embodiments, the mobile terminal 300 or other electronic devices may also simulate the functions of the control device 100 by running an application program for controlling the display device 200 .

[0089] like Figure 1 As also shown in FIG. 4 , the display device 200 also communicates data with the server 400 through various communication methods. The display device 200 may be allowed to communicate and connect through a local area network (LAN), a wireless local area network (WLAN), and other networks.

[0090] The display device 200 may provide a broadcast receiving television function, and may also additionally provide an intelligent network television function with a computer support function, including but not limited to network television, smart television, Internet Protocol television (IPTV), and the like.

[0091] Figure 2 Some embodiments of the present application provide Figure 1 2 is a block diagram of the hardware configuration of the display device 200.

[0092] In some embodiments, the display device 200 may include at least one of a tuner 210, a communication device 220, a detector 230, a device interface 240, a controller 250, a display 260, an audio output device 270, a memory, a power supply, and a user input interface.

[0093] In some embodiments, the detector 230 is used to collect signals of the external environment or external interaction. For example, the detector 230 includes a light receiver, a sensor for collecting the intensity of ambient light; or, the detector 230 includes an image collector, such as a camera, which can be used to collect external environment scenes, user attributes or user interaction gestures; or, the detector 230 includes a sound collector, such as a microphone, etc., for receiving external sounds.

[0094] In some embodiments, the display 260 includes a display function component for presenting a picture, and a driving component for driving an image display. The display 260 is used to receive an image signal output from the controller 250 for display. For example, the display 260 can be used to display video content, image content, and components of a menu control interface and a user control UI interface.

[0095] In some embodiments, the communication device 220 is a component for communicating with an external device or server 400 according to various communication protocol types. The display device 200 may be provided with a plurality of communication devices 220 according to different supported communication modes. For example, when the display device 200 supports wireless network communication, the display device 200 may be provided with a communication device 220 including a WiFi function. When the display device 200 supports Bluetooth connection communication, the display device 200 needs to be provided with a communication device 220 including a Bluetooth function.

[0096] The communication device 220 can enable the display device 200 to communicate with the external device or server 400 by wireless or wired connection. Among them, the wired connection can connect the display device 200 with the external device through components such as data cables and interfaces. The wireless connection can connect the display device 200 with the external device through wireless signals or wireless networks. The display device 200 can establish a connection relationship with the external device directly, or indirectly establish a connection relationship through a gateway, a router, a connection device, etc.

[0097] In some embodiments, the controller 250 may include at least one of a central processing unit, a video processor, an audio processor, a graphics processor, and a power processor, and a first interface to an nth interface for input / output. The controller 250 controls the operation of the display device and responds to the user's operation through various software control programs stored in the memory. The controller 250 controls the overall operation of the display device 200.

[0098] In some embodiments, the controller 250 and the tuner-demodulator 210 may be located in different separate devices, that is, the tuner-demodulator 210 may also be located in an external device of the main device where the controller 250 is located, such as an external set-top box.

[0099] In some embodiments, the user may input a user command through a graphical user interface (GUI) displayed on the display 260 , and the user input interface receives the user input command through the graphical user interface (GUI).

[0100] In some embodiments, the audio output device 270 may be a local speaker of the display device 200, or may be an external audio output device of the display device 200. In particular, for the external audio output device of the display device 200, the display device 200 may also be provided with an external audio output terminal, and the audio output device may be connected to the display device 200 through the external audio output terminal to output the sound of the display device 200.

[0101] In some embodiments, the user input interface 280 may be used to receive instructions from a user.

[0102] In order to perform user interaction, in some embodiments, the display device 200 may run an operating system. The operating system is a computer program for managing and controlling hardware resources and software resources in the display device 200. The operating system can control the display device to provide a user interface, for example, the operating system can directly control the display device to provide a user interface, or can provide a user interface by running an application program. The operating system also allows the user to interact with the display device 200.

[0103] It should be noted that the operating system may be a native operating system based on a specific operating platform, or a third-party operating system deeply customized based on a specific operating platform, or an independent operating system specially developed for the display device.

[0104] The operating system can be divided into different modules or layers according to the functions implemented, such as Figure 3 As shown, in some embodiments, the system is divided into four layers, from top to bottom, namely, the application layer (Applications) layer (referred to as "application layer"), the application framework layer (Application Framework) layer (referred to as "framework layer"), the system library layer and the kernel layer.

[0105] In some embodiments, the application layer is used to provide services and interfaces for applications so that the display device 200 can run applications and interact with users based on the applications. At least one application can be run in the application layer, and these applications can be window programs, system settings programs, clock programs, etc. that come with the operating system; they can also be applications developed by third-party developers. In specific implementations, the application packages in the application layer are not limited to the above examples.

[0106] The framework layer provides application programming interfaces (APIs) and programming frameworks for applications. The application framework layer includes some predefined functions. The application framework layer is equivalent to a processing center that determines the actions that applications in the application layer take. Through the API interface, applications can access system resources and obtain system services during execution.

[0107] like Figure 3 As shown, in the embodiment of the present application, the application framework layer includes a view system, managers, content providers, etc., wherein the view system can design and implement the interface and interaction of the application, and the view system includes lists, grids, text boxes, buttons, etc. The manager includes at least one of the following modules: an activity manager for interacting with all activities running in the system; a location manager for providing system services or applications with access to system location services; a package manager for retrieving various information related to the application package currently installed on the device; a notification manager for controlling the display and clearing of notification messages; and a window manager for managing icons, windows, toolbars, wallpapers, and desktop widgets on the user interface.

[0108] In some embodiments, the activity manager is used to manage the life cycle of each application and the usual navigation back function, such as controlling the exit, opening, and back of the application. The window manager is used to manage all window programs, such as obtaining the display screen size, determining whether there is a status bar, locking the screen, capturing the screen, and controlling the display window changes, for example, reducing the display window, shaking the display, distorting the display, etc.

[0109] In some embodiments, the system runtime layer can provide support for the framework layer. When the framework layer is used, the operating system will run the instruction library contained in the system runtime layer, such as the C / C++ instruction library, to implement the functions to be implemented by the framework layer.

[0110] In some embodiments, the kernel layer is a functional layer between the hardware and software of the display device 200. The kernel layer can implement functions such as hardware abstraction, multitasking, and memory management. Figure 3 As shown, the kernel layer may be configured with hardware drivers, and the drivers included in the kernel layer may be at least one of the following drivers: audio driver, display driver, Bluetooth driver, camera driver, WIFI driver, USB driver, HDMI driver, sensor driver (such as fingerprint sensor, temperature sensor, pressure sensor, etc.), and power driver, etc.

[0111] It should be noted that the above example is only a simple division of the operating system functions and does not constitute a limitation on the specific operating system form of the display device 200 in the embodiment of the present application. Depending on factors such as the function of the display device and the type of operating system, the number of levels and specific level types contained in the operating system may be expressed in other forms.

[0112] The display device 200 may also be a display device. In some embodiments, the display device 200 may also have a voice interaction function. A user may send a voice command to the display device 200 to control the display device 200 to perform an interactive action corresponding to the voice command, thereby improving the portability of the display device 200.

[0113] In order to facilitate the user to use the voice interaction function of the display device 200, the display device 200 can also integrate a voice application, which can quickly recognize the voice commands sent by the user to improve the interaction efficiency. The display device 200 can integrate a voice application on the display device 200 through a software development kit (SDK). The voice application is an application that implements the voice interaction function through voice recognition and natural language processing technology. When the user sends a voice command to the display device 200, the voice application can recognize the voice command, thereby controlling the display device 200 to perform the corresponding functional operation, such as searching for programs, playing music, querying information, etc.

[0114] When receiving a voice command, the display device 200 may parse the voice command to obtain the interactive action that the user wants to control the display device 200 to perform according to the analysis result. For example, if the voice command is "open the movie played by character A", the display device 200 may parse the keywords in the voice command and perform the interactive action according to the keywords or keyword combinations obtained by the analysis. For example, the display device 200 may parse the keywords "character A" and "movie" in the voice command and combine the keywords to obtain the final search keyword "character A+movie", thereby searching for the movie media corresponding to character A according to the search keyword, and playing it through the display 260 to complete the voice control interaction.

[0115] In order to improve the efficiency of parsing voice instructions, the display device 200 can call the label template in the label template library to parse the voice instructions. The label template can include a structure composed of different category tags. For example, the label template can be "character label + format label". The display device 200 can match the keywords in the voice instruction according to the different category labels, so as to parse the voice instruction into a machine instruction that can be executed by the display device 200 according to the matching result, and complete the parsing process. For example, voice instruction 1 "Open songs sung by singer B", or voice instruction 2 "Search for books written by writer C", etc., are all combinations of characters and media formats. Therefore, the display device 200 can parse the voice instruction 1 into a play instruction of "singer B + song" through "character label + format label", and the display device 200 can parse the voice instruction 2 into a search instruction of "writer C + book" through "character label + format label".

[0116] The label template needs to be manually written, and the display device 200 can collect the completed label templates into a label template library. However, corresponding to a large number of vocabulary combinations, the amount of data of the label template to be written is large, resulting in a large amount of time consumed in the writing process, reducing the generation efficiency of the label template library.

[0117] In order to solve the problem of time-consuming manual writing of label templates, some embodiments of the present application provide a display device 200, including a display, configured to display a user interface; a memory, configured to store test data, the test data is text data containing preset text content, and is used to generate a label template; a controller 250, configured to execute a label template library generation method. Figure 4 A flowchart of a method for generating a label template library executed by a display device provided in an embodiment of the present application. Figure 5 The timing diagram of the display device provided in the embodiment of the present application executing the label template library generation method. The display device provided in the embodiment of the present application executes the label template library generation method according to the following steps: Figure 5 The timing relationship shown in the figure executes the label template library generation method. Figure 4 and Figure 5 , the method includes the following:

[0118] S100: In response to a control instruction for constructing a label template library, obtain test data.

[0119] The process of constructing the label template can be carried out when the display device 200 is in the test stage or when the display device 200 is in the application stage. In the test stage, the controller 250 can, in response to a control instruction for constructing a label template library, obtain test data. Among them, the controller 250 can include various ways of obtaining test data. For example, the controller 250 can receive test data sent by a user or other devices, or can obtain test data through the Internet when the display device 200 is in a network-connected state.

[0120] In some embodiments, the test data can include text data in multiple different fields to improve the generality of the label template. For example, the test data can include the children's field, the learning field, the sports field, the entertainment field, or the game field, etc. For the display device 200 used in a specific field, the test data can also be concentrated in that specific field.

[0121] S200: Perform word segmentation processing on the test data to obtain a word segmentation result.

[0122] After obtaining the test data, in order to improve the accuracy of processing the test data, the controller 250 can perform word segmentation processing on the test data and assign category labels to the obtained word segments. For this purpose, the controller 250 can include a word segmentation module and a labeling module. Among them, the controller 250 can perform text cleaning on the test data through the word segmentation module to remove irrelevant information in the test data and improve the text quality of the test data. Among them, the irrelevant information is information that has nothing to do with the text semantics of the test data. For example, punctuation marks, spaces, and also stop words such as "and", "is", "of", etc. The word segmentation module can include the Jieba word segmenter. During the process of word segmentation, the controller 250 can perform word segmentation processing on the test data through the Jieba word segmenter of the word segmentation module.

[0123] After completing the text cleaning, the controller 250 can perform word segmentation processing on the test data to divide the test data into multiple word segmentation results. Among them, the word segmentation result can include the word segment and the category label corresponding to the word segment. The controller 250 can analyze the word segment semantics of each word segment and assign a category label to the word segment according to the word segment semantics. The category label is used to represent the entry category corresponding to the word segment. For example, as Figure 6As shown, in the map navigation function, the test data may be "query the route from city A to city B". The controller 250 performs word segmentation processing on the test data, and the word segments obtained are "query", "city A", "city B" and "route" in sequence, wherein the controller 250 may parse the semantics of the word segments and assign category labels to the word segments according to the semantics of the word segments, for example, "query" is used to instruct the display device 200 to perform a query action, therefore, the controller 250 may assign an action label or an execution label to "query", "city A" and "city B" are city names, respectively, and the controller 250 may assign city labels to "city A" and "city B", respectively, and "route" is the final information format, therefore, the controller 250 may assign a format label to "route", etc.

[0124] In some embodiments, the controller 250 may search for the segmented words to determine the meaning of the segmented words, and then assign labels to the segmented words through a labeling module according to the meaning of the segmented words to obtain the segmented results.

[0125] S300: Sort the category labels to obtain a label template.

[0126] The test data may include multiple word segmentation results. Since each word segmentation result includes a category label, a label template may be generated according to the category label. The controller 250 should generate the label template in the order of word segmentation results obtained by performing word segmentation processing on the test data. To this end, the controller 250 may obtain the word segmentation order of the word segmentation results, and sort the category labels according to the word segmentation order, thereby generating the label template according to the sorting result of the category labels.

[0127] For example, Figure 7 As shown, the test data is "movies starring characters A and B". According to the order of word segmentation processing, i.e., the word segmentation order, the controller 250 can perform word segmentation processing on the test data to obtain the word segments: "character A", "character B" and "movie". The category label of "character A" is the character label, the category label of "character B" is the character label, and the category label of "movie" is the format label. According to the word segmentation order, the sorting result of the category labels is: character label-character label-format label. The controller 250 can generate a label template "character label + character label + format label" according to the results of the category labels.

[0128] In some embodiments, the category tag may also include multiple subcategory tags to refine the field of tag classification. For example, in the "character tag", according to the different character attributes corresponding to the segmentation, it can also be divided into "actor tag", "singer tag", "author tag", etc. The controller 250 can also replace the category tag with a subcategory tag. For example, for the segmentation of "singer", the controller 250 can assign it a character tag or a singer tag. When replacing the assignment of the category tag with a subcategory tag, the controller 250 can also mark the source of the subcategory tag, for example, the singer tag belongs to the character tag.

[0129] During the label assignment process performed by the controller 250, the controller 250 may assign two or more different category labels to the segmented words according to different categories of the segmented words in the same field. Different category labels may be derived from the same category and assigned in the form of subcategory labels. For example, Figure 8 As shown, for "character A", who is both an actor and a singer, the controller 250 can assign both an actor tag and a singer tag to "character A". At this time, the category tag of "character A" can be a category tag of "actor tag AND singer tag", or the actor tag and the singer tag can be directly assigned to "character A".

[0130] In the above example, the actor label and the singer label are both character labels and belong to the same category. In the actual process of assigning labels, multiple different category labels can be assigned to the segmented words according to their different meanings and attributes.

[0131] For word segments assigned with multiple category tags, the multiple category tags can be merged first to obtain a merged tag, and then a tag template is generated based on the category tag and the merged tag according to the word segmentation order. Continuing with the above example, when "character B" is assigned with both an actor tag and a singer tag, the category tag of "character B" can be merged into "actor tag AND singer tag", and correspondingly, the tag template generated by the controller 250 according to the sorting result of the category tags is "actor tag + actor tag AND singer tag + format tag".

[0132] It should be noted that in the above embodiments in which the category labels are described in detail, the segmented words may be specific word names, for example, "city A" may be a specific city name, and "person A" may be a specific person name, etc.

[0133] From the above technical features, it can be seen that the classification field of category labels is refined by subcategory labels, thereby increasing the diversity of label categories and improving the accuracy of voice command recognition.

[0134] S400: Perform clustering on the label templates to obtain at least one clustering template set.

[0135] Since different word segments have different category labels, the resulting label templates are also different. In order to improve the efficiency of voice command recognition, the controller 250 can classify the label templates and aggregate the label templates into template sets according to different template categories.

[0136] Corresponding to the functions that the display device 200 can execute, the template category may include a media resource search category, such as "search for children's channels", "search for movies of ×××", "search for songs of ×××", etc. The label template of the media resource search category is used to instruct the display device 200 to search for the corresponding content and play or display it. The template category may also include a function category, such as "increase the volume", "turn off the power", "turn on the network function", etc. The label template of the function category is used to instruct the display device 200 to execute the corresponding function. The above is only an exemplary description of two categories, and the controller 250 can set the corresponding label template according to the executable functions of the display device 200.

[0137] In some embodiments, the label templates may be classified by clustering to cluster and merge label templates of the same category to obtain a cluster template set. In the same cluster template set, all label templates are of the same category.

[0138] Fig. 9 The flow chart of the display device provided by the present application performing cluster merging according to the label template. Fig. 9 During the clustering process, the controller 250 can obtain the first similarity between different label templates, and determine whether the label templates are of the same category by setting a similarity threshold. If the first similarity is greater than or equal to the similarity threshold, it means that the compared label templates are of the same category, and the controller can cluster and merge the label templates. During the merging process, a target label template can be selected from the label templates corresponding to the first similarity, and a clustering label can be generated based on the template structure of the target label template. For example, if the target label template is "category label A+category label B+category label B", the controller 250 can obtain the label structure of the target label template as an "ABB" structure, thereby generating a clustering label for the clustering template set based on the "ABB" structure.

[0139] The cluster label is a collection label of the cluster template set. The controller 250 can output the merged label template as a cluster template set according to the cluster label. If the first similarity is less than the similarity threshold, it means that the differences between the label templates for similarity comparison are large and cannot be clustered into the same label category. The controller 250 can add other label templates to continue calculating the first similarity between different label templates.

[0140] The controller 250 can calculate the first similarity between different label templates based on the category relationship of the category tags at corresponding positions in different label templates. When the category tags at corresponding positions in different category tags are of the same category, the first similarity can be improved. When the category tags at corresponding positions in different category tags are all different, or one of them is different, the first similarity will be reduced. Alternatively, although the category tags between the label templates are of the same category, when the same category tags are located in different positions in their respective label templates, the first similarity will also be reduced.

[0141] In some embodiments, the controller 250 can set the number of label templates to be merged. For example, when the number of merges is 3, the controller 250 can simultaneously calculate the first similarity between the three label templates. If the first similarity is greater than or equal to the similarity threshold, the three label templates are clustered and merged at the same time. The controller 250 can also change the number of merges in response to an instruction to adjust the number of merges, thereby improving clustering accuracy.

[0142] It can be seen from the above technical solutions that by setting the number of label templates to be merged, multiple label templates can be clustered and merged at the same time, thereby accelerating the merging rate of label templates and improving the efficiency of clustering and merging.

[0143] In some embodiments, the controller 250 may perform a clustering and merging process in the memory. After the controller 250 merges some label templates, the memory includes the label templates that have not been merged and the clustering template set. Therefore, the controller 250 needs to continue to perform the clustering process to merge all label templates into the clustering template set of the corresponding category. Fig.10 As shown, the controller 250 can also obtain a second similarity between the label template and the label template in the cluster template set. The controller 250 can continue to use the similarity threshold as the judgment threshold for determining whether to perform cluster merging of the label template and the cluster template set. When the second similarity is greater than or equal to the similarity threshold, it means that the label template and the label template in the cluster template set are of the same category, and the controller 250 can merge the label template into the cluster template set. When the second similarity is less than the similarity threshold, it means that the label template and the label template in the cluster template set are of different categories, and the controller 250 can replace the label template to calculate the second similarity between the replaced label template and the label template in the cluster template set.

[0144] The second similarity is calculated in a similar manner to the first similarity, except that the controller 250 calculates the label template of the uncompleted clustering and the label template of the completed clustering merge. For ease of description, the label template of the completed clustering merge can be defined as a cluster template. The controller 250 can calculate the category relationship between the category labels at the corresponding positions of the label template and the cluster template.

[0145] Correspondingly, the controller 250 can also calculate the third similarity between the cluster template sets during the cluster merging process. For example, for the first cluster template set and the second cluster template set, the controller 250 can calculate the third similarity between the first cluster label and the second cluster label, and also use the similarity threshold as the judgment threshold. If the third similarity is greater than or equal to the similarity threshold, the controller 250 performs cluster merging on the first cluster template set and the second cluster template set, and selects a cluster label from the first cluster label and the second cluster label as the merged cluster template set.

[0146] In some embodiments, the display device 200 can also increase or decrease application functions by installing or uninstalling application programs during the application process. Therefore, the controller 250 can also detect the application functions that the display device 200 can execute, and thus increase or decrease the label templates according to the increase or decrease of the application functions. For example, when the display device 200 installs a voice conversion application, that is, a new voice conversion function is added, the controller 250 can obtain new test data related to the voice function, and perform word segmentation processing on the new test data to obtain new category labels related to the voice conversion function, and generate new label templates suitable for the voice conversion function based on the new category labels. When the display device 200 reduces the application function, the controller 250 can remove the label template of the corresponding function from the label template library to improve the efficiency of parsing voice commands.

[0147] S500: Matching the semantic information of the test data with the set label corresponding to the clustering template set to obtain the matching probability between the clustering template set and the test data.

[0148] After cluster merging is performed, multiple cluster template sets will be generated. Fig.11 As shown, the controller 250 can perform a test check on the clustering template set through the test data, that is, the controller 250 matches the semantic information of the test data with the clustering labels corresponding to the clustering template set to obtain the matching probability between the clustering template set and the test data. In this process, the controller 250 can parse the semantic information of the test data, thereby matching the clustering labels according to the semantic information to obtain the number of label matches of the clustering labels.

[0149] The controller 250 also needs to obtain the number of matches of the label templates in each cluster template set whose semantic similarity with the semantic information of the test data is greater than or equal to the semantic similarity threshold and as the total number of matches, so as to calculate the total number of matches of the test data with all the cluster labels in the test verification and calculate the matching probability. For example, the number of first label matches of the test data with the first cluster label is 4 times, and the total number of matches of the test data with all the cluster labels in the test verification is 10 times, then the matching probability of the first cluster template set is 40%.

[0150] If the test data successfully matches the cluster label, it means that there is a label template in the cluster template set that matches the test data. Therefore, the controller 250 can further match the test data according to the label template in the cluster template set. The controller 250 can obtain the semantic information of the test data, and match the semantic information according to the label template to obtain a matching result. If the semantics of the label template and the semantic information are the same, or the semantic similarity between the label template and the semantic information is greater than or equal to the semantic similarity threshold, the label template successfully matches the test data.

[0151] The controller 250 may record the first matching number of the label template and the test data during the test verification process. During the matching process, the controller 250 needs to match different test data through the label template, so as to continuously update the first matching number of the label template according to the matching result. The controller 250 may sum each of the first matching times to obtain the number of label matchings in which the clustering label matches the semantic information of the test data.

[0152] Based on this, the controller 250 can also calculate the matching probability of the cluster template set according to the first matching number of the label templates in the cluster template set. After completing the test verification process, the controller 250 can obtain the total matching number of the test data and all label templates. The cluster template set may include at least two label templates. Therefore, the matching probability of the test data and the cluster template set can be calculated based on the matching probability of all label templates in the cluster template set, that is, the matching probability of the cluster template set is calculated based on the sum of the first matching numbers of all label templates in the cluster template set, that is, the label matching number and the total matching number.

[0153] For example, the first cluster template set includes the first label template and the second label template, wherein the first matching number of the first label template is 1, the first matching number of the second label template is 3, and the total matching number of the first label template and the second label template is 4, that is, the number of label matching times of the first cluster template set is 4. When the total number of matching times of the test data for all cluster labels is 10 times, the matching probability of the first cluster template set is 40%.

[0154] S600: Filter out a first template set whose matching probability is greater than or equal to a probability threshold from the at least one clustering template set.

[0155] The controller 250 can set a matching probability threshold and classify the cluster template set according to the matching probability threshold. If the matching probability is greater than or equal to the matching probability threshold, the cluster template set is marked as the first template set. If the matching probability is less than the matching probability threshold, the cluster template set is marked as the second template set, and the second template set is a cluster template set with a lower response rate. In order to save the matching efficiency of the display device 200, the controller 250 can filter out the second template set to retain the first template set with a relatively high matching probability.

[0156] Based on this, the controller 250 filters out the second template set with a lower matching probability and retains the first template set with a relatively higher matching probability to improve the coverage of the template set in response to voice commands during the application phase of the display device 200.

[0157] In some embodiments, the controller 250 may also calculate the matching probability of the label template based on the first matching number and the total matching number of the single label template. The controller 250 may set a template probability determination threshold. When the matching probability of the label template is greater than or equal to the template probability determination threshold, the label template may be marked as the first template. When the matching probability of the label template is less than the template probability determination threshold, the label template may be marked as the second template, and the second template may be removed from the cluster template set to retain the first template.

[0158] S700: Generate a label template library according to the first template set.

[0159] When filtering out the low-frequency template set in the cluster template set, the remaining first template set is a cluster template set with a higher coverage rate for voice commands. After obtaining the first template set, the controller 250 can classify and aggregate the first template set to generate a label template library.

[0160] In some embodiments, it is considered that when a user inputs a voice command, the voice command may be input in a different word order. For example, "open the settings interface" and "open the settings interface", the above two semantics and the execution effect when the display device 200 responds to the voice command are the same, the only difference is the order of appearance of the words. Since the label template is generated according to the category label corresponding to the word segmentation order, the word segmentation order obtained by the controller 250 is not the same for different word orders, and the corresponding label template is also different. When the word orders of two test data are different, but the semantics are the same, the controller 250 will only effectively match the label template with one of the test data according to the word order, that is, the test data with the same word order and category label order, thereby causing the other test data to fail to match and reduce the matching probability.

[0161] Therefore, in order to improve the calculation accuracy of matching efficiency, such as Fig.13 As shown, the controller 250 can also change the arrangement order of the category labels in the label template during the test and verification process to obtain a random label template. The random label template can be in reverse order or random order to disrupt the category labels in the label template, so as to match the random label template with the test data again to obtain a second matching number of the random label template and the test data.

[0162] The above technical features can accurately match test data with different word orders but the same semantics by changing the order of category labels in the label template, thereby improving the matching probability of the label template.

[0163] The random label template and the original label template are only changes in the order of category labels. The interactive effect of matching the test data with the random label template and matching the test data with the original label template is the same. For this reason, the controller 250 can update the first matching number of the original label template according to the second matching number of the random label template, and calculate the matching probability based on the updated first matching number and the total matching number.

[0164] In some embodiments, after generating a label template, the controller 250 may also detect a format tag of the label template, where the format tag is a category tag in the label template used to represent format information. For example, "play songs", "query routes", "search information". In label templates such as "open functions", "songs", "routes", "information" and "functions" are the format information of label templates. The controller 250 may determine the interaction effect corresponding to the label template based on the format tag, so that when the voice command matches the label template, the corresponding interaction effect is executed according to the label template.

[0165] If no format tag is detected in the tag template, the controller 250 cannot determine the interactive effect of the tag template. For example, the test data "play character A" corresponds to the word segmentation "play" and "character A", and the corresponding category tags are "action tag" and "character tag", that is, the tag template "action tag + character tag". The controller 250 cannot determine the format information of the media to be played according to the tag template.

[0166] Therefore, when no format tag is detected in the label template, the controller 250 can generate a format tag based on other category tags in the label template. For example, when the character tag is a singer tag, the controller 250 can determine that the word "character A" is a singer. Therefore, the controller 250 can generate a song tag as a format tag based on the singer tag, and add the format tag to the corresponding position of the label template, that is, "action tag + character tag + format tag (song tag)".

[0167] In some embodiments, if a format tag is detected in a tag template, a strong rule tag may be added to the format tag, the strong rule tag being used to indicate that the format tag should not be deleted, thereby avoiding the problem of deleting the format tag due to a low matching probability of the format tag.

[0168] In some embodiments, during the test verification process, the controller 250 can detect invalid test data that is not successfully matched. Invalid test data is test data that does not match any label template during the test verification process, that is, invalid test data is test data whose semantic information has a semantic similarity with all label templates that is less than the semantic similarity.

[0169] The controller 250 may repeat the steps of generating the label template for the invalid test data. In this process, the controller 250 may perform the word segmentation process on the invalid test data again to obtain an iterative word segmentation result, and the iterative word segmentation result includes an iterative word segmentation and an iterative category label corresponding to the iterative word segmentation. In order to distinguish it from the word segmentation result of the first iteration, the controller 250 changes the position of the word segmentation during the iterative word segmentation process, thereby obtaining an iterative word segmentation different from the first word segmentation result. After obtaining the iterative word segmentation result, the controller 250 may generate an iterative label template according to the iterative category label corresponding to the iterative word segmentation result.

[0170] In some embodiments, the controller 250 traverses the cluster template set according to the test data to determine the cluster label that matches the test data. If the number of cluster template sets is too large, the matching efficiency of the test data will be affected. To this end, the controller 250 can set the number of clusters to limit the number of cluster template sets, and the number of clusters is the maximum number of cluster template sets. Fig.14As shown, the controller 250 may monitor the number of cluster template sets. When the number of cluster template sets reaches the number of clusters, the maximum number of cluster template sets has been reached, so the controller 250 may stop merging label template clusters into new cluster template sets.

[0171] Stopping the merging to generate a new cluster template set will not affect the controller 250 merging the label template into the already generated cluster template set. After all label templates are merged into the cluster template set of the corresponding category, if there are remaining label templates, the controller 250 can generate a cluster template set based on the remaining label templates. However, the cluster template set generated by the remaining label templates will also occupy a set number. Therefore, the controller 250 can stop clustering and merging label templates into a new cluster template set when the difference between the number of set clusters and the number of cluster template sets is 1. After generating a cluster template set with the remaining label templates, the number of cluster template sets is equal to the number of set clusters, that is, the maximum number of cluster template sets is reached.

[0172] Some embodiments of the present application also provide a label template library generation method, which is applied to a display device 200, wherein the display device 200 includes a display 260, a memory, and a controller 250, wherein the display 260 is configured to display a user interface; the memory is configured to store test data; the method includes:

[0173] S100: Responding to a control instruction for building a label template library, acquiring test data;

[0174] S200: Perform word segmentation processing on the test data to obtain a word segmentation result.

[0175] The word segmentation results include the word segmentation and the category label corresponding to the word segmentation.

[0176] S300: Sort the category labels to obtain a label template.

[0177] S400: performing clustering on the label templates to obtain at least one clustering template set.

[0178] S500: Matching the semantic information of the test data with the set label corresponding to the clustering template set to obtain a matching probability between the clustering template set and the test data.

[0179] The matching probability is the probability that the clustering labels of the clustering template set match the semantic information of the test data.

[0180] S600: Filter out a first template set whose matching probability is greater than or equal to a probability threshold from the at least one clustering template set.

[0181] The first template set is a clustering template set whose matching probability is greater than or equal to a probability threshold, and the clustering label is a set label of the clustering template set.

[0182] S700: Generate the label template library according to the first template set.

[0183] It can be seen from the above technical solutions that the present application provides a display device and a label template library generation method, which obtains test data in response to a control instruction for building a label template library, performs word segmentation processing on the test data, and obtains a word segmentation result with a category label. Generate a label template according to the category label, and cluster the label template to obtain a cluster template set. Perform a test check on the cluster template set according to the test data, and filter out a first template set to generate a label template library. The present application solves the problem of time-consuming manual writing and low efficiency in generating a label template library by performing word segmentation on the test data, generating a label template according to the word segmentation result, and clustering the label template to generate a label template library according to the cluster template set.

[0184] The same and similar parts between the various embodiments in this specification can be referenced to each other and will not be described again here.

[0185] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution in the embodiments of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of various embodiments of the present invention or certain parts of the embodiments.

[0186] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0187] For the convenience of explanation, the above description has been made in conjunction with specific embodiments. However, the above exemplary discussion is not intended to be exhaustive or limit the embodiments to the specific forms disclosed above. Based on the above teachings, various modifications and variations can be obtained. The selection and description of the above embodiments are to better explain the principles and practical applications, so that those skilled in the art can better use the embodiments and various different variations of the embodiments suitable for specific use considerations.

Claims

1. A display device, characterized in that: include: a display configured to display a user interface; a memory configured to store test data; The controller is configured as: Responding to a control instruction for building a label template library, acquiring the test data; Performing word segmentation processing on the test data to obtain a word segmentation result, wherein the word segmentation result includes a word segmentation and a category label corresponding to the word segmentation; Sorting the category labels to obtain a label template; Performing clustering on the label template to obtain at least one cluster template set, wherein the cluster template set has corresponding cluster labels; Matching the semantic information of the test data with the clustering labels corresponding to the clustering template set to obtain a matching probability between the clustering template set and the test data; Filtering out a first template set whose matching probability is greater than or equal to a probability threshold from the at least one clustering template set; The label template library is generated according to the first template set.

2. The display device according to claim 1, characterized in that The controller performs the step of clustering the label templates to obtain at least one clustering template set, which is specifically configured as follows: Calculating first similarities between different label templates according to the category relationship between the corresponding positions of the category labels in the label templates; If the first similarity is greater than or equal to a similarity threshold, selecting a target label template from label templates corresponding to the first similarity; Generate a clustering label according to the template structure of the target label template; Clustering and merging the label templates to obtain the clustering template set; Add the corresponding clustering label to the clustering template set.

3. The display device according to claim 2, characterized in that The controller performs the step of clustering and merging the label templates to obtain the cluster template set, which is specifically configured as follows: Calculating a second similarity according to the category relationship between the category labels at corresponding positions in the label template and the cluster template, wherein the cluster template is a label template that has completed cluster merging; If the second similarity is greater than or equal to the similarity threshold, the label template is merged into the clustering template set.

4. The display device according to claim 1, characterized in that The controller performs the step of sorting the category labels to obtain a label template, which is specifically configured as follows: Obtaining the word segmentation order of the word segmentation corresponding to each of the word segmentation results; According to the word segmentation order, the category labels in the corresponding word segmentation results are sorted to obtain a sorting result; A label template is generated according to the sorting result.

5. The display device according to claim 4, characterized in that The controller performs the step of matching the semantic information of the test data with the clustering labels corresponding to the clustering template set, and is specifically configured as follows: Obtaining a first matching number of the label templates in the clustering template set whose semantic similarity with the semantic information of the test data is greater than or equal to a semantic similarity threshold; Summing each of the first matching times to obtain the number of label matches where the clustering label matches the semantic information of the test data; Obtaining the total number of matches of the label templates in each of the clustering template sets, whose semantic similarity with the semantic information of the test data is greater than or equal to a semantic similarity threshold; The matching probability between the test data and the clustering template set is calculated according to the number of label matches and the total number of matches.

6. The display device according to claim 5, characterized in that The controller is also configured to: Changing the order of the category labels in the label template to obtain a random label template; Obtaining a second matching number of the out-of-order label template that matches the test data; The first matching number is updated according to the second matching number.

7. The display device according to claim 6, characterized in that After the controller executes the step of changing the arrangement order of the category labels in the label template, the controller is further configured to: Detecting a format tag of the label template, where the format tag is a category tag in the label template used to represent format information; If no format tag is detected in the tag template, generating a format tag according to other category tags in the tag template; Add the format tag in the tag template.

8. The display device according to claim 5, characterized in that After the controller performs the step of matching the semantic information of the test data with the clustering labels corresponding to the clustering template set, the controller is further configured to: Obtain invalid test data from each of the test data, wherein the invalid test data is test data whose semantic similarity between semantic information and the label template is less than a semantic similarity threshold; Performing iterative word segmentation on the invalid test data to obtain an iterative word segmentation result, wherein the iterative word segmentation result includes an iterative word segmentation and an iterative category label corresponding to the iterative word segmentation; Generate an iteration label template based on the iteration category label.

9. The display device according to claim 2, characterized in that: The controller performs the step of clustering and merging the label templates, and is specifically configured as follows: Obtaining the number of clusters, where the number of clusters is the maximum number of the clustering template set; In the process of clustering and merging the label templates, obtaining the number of the cluster template sets; When the number of clustering template sets is equal to the number of clusters, the label template clustering and merging into a new clustering template set is stopped.

10. A method for generating a label template library, characterized in that: Applied to a display device, the display device comprises a display, a memory and a controller, the display being configured to display a user interface; The memory is configured to store test data; the method comprises: Responding to a control instruction for building a label template library, acquiring the test data; Performing word segmentation processing on the test data to obtain a word segmentation result, wherein the word segmentation result includes a word segmentation and a category label corresponding to the word segmentation; Sorting the category labels to obtain a label template; Performing clustering on the label template to obtain at least one cluster template set, wherein the cluster template set has corresponding cluster labels; Matching the semantic information of the test data with the clustering labels corresponding to the clustering template set to obtain a matching probability between the clustering template set and the test data; Filtering out a first template set whose matching probability is greater than or equal to a probability threshold from the at least one clustering template set; The label template library is generated according to the first template set.

Images