WO2021120875A1

WO2021120875A1 - Search method and apparatus, terminal device and storage medium

Info

Publication number: WO2021120875A1
Application number: PCT/CN2020/124762
Authority: WO
Inventors: 彭璐; 赵安; 于超
Original assignee: 华为技术有限公司
Priority date: 2019-12-20
Filing date: 2020-10-29
Publication date: 2021-06-24
Also published as: CN111125523A; CN111125523B

Abstract

An artificial intelligence (AI) based search method and apparatus, a terminal device and a storage medium, wherein same relate to the technical field of computers. The method comprises: acquiring a search keyword (S101); according to a pre-established mapping relationship list regarding search objects and search keywords, determining a target search object corresponding to the search keyword (S102); and displaying the target search object (S103), wherein the mapping relationship list regarding search objects and search keywords is established according to historical search behaviors and operation behaviors after the historical search behaviors. A target search object corresponding to a search keyword is determined by means of a mapping relationship list regarding search objects and search keywords, such that a more accurate search object can be recommended to a user.

Description

Search method, device, terminal equipment and storage medium

This application claims the priority of a Chinese patent application filed with the State Intellectual Property Office, the application number is 201911328504.5, and the application name is "search method, device, terminal equipment and storage medium" on December 20, 2019, the entire content of which is incorporated by reference In this application.

Technical field

This application relates to the field of computer technology, and in particular to a search method, device, terminal device, and storage medium based on artificial intelligence (AI).

Background technique

The existing terminal device recommends search results for the user after the user enters the search keyword, thereby facilitating the user to quickly find the corresponding service. However, the existing search methods mainly return search results according to the degree of matching between the name of the search object and the search keyword entered by the user. For some search objects whose names do not match the search keywords, accurate recommendations cannot be made. Therefore, the existing The search method cannot accurately determine the user’s true search intent.

Summary of the invention

The embodiments of the present application provide a search method, device, terminal device, and storage medium to accurately determine the user's real search intention.

In the first aspect, an embodiment of the present application provides a search method, including:

Get search keywords;

Determine the target search object corresponding to the search keyword according to a pre-established list of mapping relations between search objects and search keywords; wherein, the list of mapping relations between search objects and search keywords is based on historical search behaviors and the history Established by the operation behavior after the search behavior;

The target search object is displayed.

In the embodiment of this application, the mapping relationship list between search objects and search keywords is established based on historical search behaviors and operation behaviors after historical search behaviors. Since the operation behaviors after historical search behaviors reflect the user’s real search intentions, the terminal device When obtaining search keywords, determining the target search object according to the mapping relationship list between the search object and the search keyword can recommend more accurate objects for the user and improve the user experience.

In a possible implementation manner of the first aspect, the mapping relationship list between search objects and search keywords is established in the following manner:

Obtain search behavior log information and operation behavior log information of the user, where the search behavior log information includes historical search behaviors, and the operation behavior log information includes historical operation behaviors. The search keyword is obtained from the search behavior log information, the operation behavior within a preset time interval after the user inputs the search keyword is obtained from the operation behavior log information, and the operation object corresponding to the operation behavior is obtained As an initial search object; calculate the confidence score of the initial search object, where the confidence score reflects the degree of association between the initial search object and the search keyword, and establish the search object and the search keyword according to the confidence score of the initial search object The mapping relationship list of the search keywords, thereby recommending the initial search objects with high relevance to the search keywords to the user, and improving the accuracy of the recommendation.

In a possible implementation manner of the first aspect, the calculating the confidence score of the initial search object includes:

Acquire at least one behavior characteristic value of the initial search object from the operation behavior log information, where the behavior characteristic value includes the use times of the initial search object, the average use time of the initial search object, and/or the use times of the initial search object Accounted for. Calculate the confidence score of the initial search object according to the at least one behavior feature value and a preset weight coefficient corresponding to each behavior feature value. Since the behavior feature value represents the information of the user operating the initial search object, the preset weight coefficient reflects the importance of each behavior feature value. The initial search object's value is calculated by the behavior feature value and the preset weight coefficient corresponding to each behavior feature value. The confidence score reflects the user's intention of the initial search object when searching.

The confidence score of the initial search object is calculated according to the search keyword, the initial search object, and a preset prediction model, where the preset prediction model is the search keyword and the initial search keyword corresponding to the The search object and the confidence score are the training samples, which are obtained by training the learning model using machine learning or deep learning algorithms. Among them, the preset prediction model can be used repeatedly, and the preset prediction model is used to calculate the confidence score of the initial search object to ensure the stability of the calculation result.

In a possible implementation manner of the first aspect, the establishing a mapping relationship list between the search object and the search keyword according to the confidence score of the initial search object includes:

Calculating the weight value of the search keyword according to the confidence score of the initial search object;

A mapping relationship list between the search object and the search keyword is established according to the weight value of the search keyword. It can be understood that each search object corresponds to multiple search keywords, different search objects may have the same search keyword, and different search keywords may have different weight values for different search objects.

In a possible implementation manner of the first aspect, the calculating the weight value of the search keyword according to the confidence score of the initial search object includes:

Converting the confidence score of the initial search object into the confidence score of the search keyword;

The confidence score of the search keyword that is greater than the preset confidence threshold is used as the weight value of the search keyword.

Exemplarily, after the weight value of the search keyword is obtained, the search keyword corresponding to each search object is sorted in descending order according to the weight value of the search keyword, and then a preset number of records are taken to obtain the search object and the search key A list of word mapping relations.

In a possible implementation manner of the first aspect, the determining the target search object corresponding to the search keyword includes:

The search object corresponding to the search keyword with the weight value greater than the preset weight threshold is used as the target search object, for example, the search objects are arranged in descending order according to the weight value, and the search objects after the arrangement are displayed on the terminal device.

In a possible implementation manner of the first aspect, before the obtaining the search keyword from the search behavior log information, the method further includes:

Preprocessing the search behavior log information and the operation behavior log information respectively to obtain preprocessed target search behavior log information and target operation behavior log information;

Correspondingly, obtaining search keywords from the search behavior log information, and obtaining operation behaviors within a preset time interval after the user inputs the search keywords from the operation behavior log information includes:

The search keywords are obtained from the target search behavior log information, and the operation behaviors within the preset time interval after the user inputs the search keywords are obtained from the target operation behavior log information, so as to obtain effective log information and ensure Accuracy of calculation results.

In a possible implementation of the first aspect, after the obtaining the operation behavior within a preset time interval after the user inputs the search keyword from the operation behavior log information, the method further includes:

Delete operation behaviors whose operation frequency is greater than the first threshold and delete operation behaviors whose operation frequency is less than the second threshold from the acquired operation behaviors to obtain effective operation behaviors;

Correspondingly, using the operation object corresponding to the operation behavior as the initial search object includes:

The operation object corresponding to the effective operation behavior is used as the initial search object, thereby preventing interference of high-frequency operation and low-frequency operation on the calculation result, and ensuring the accuracy of the calculation result.

In the second aspect, an embodiment of the present application provides a search device, including:

Obtaining module for obtaining search keywords;

The determining module is used to determine the target search object corresponding to the search keyword according to a pre-established list of mapping relations between search objects and search keywords; wherein, the list of mapping relations between search objects and search keywords is based on historical search Behaviors and operational behaviors after the historical search behavior;

The display module is used to display the target search object.

In a possible implementation of the second aspect, the search device further includes a mapping relationship establishment module, and the mapping relationship establishment module includes:

The obtaining unit is used to obtain the user's search behavior log information and operation behavior log information;

The extraction unit is configured to obtain search keywords from the search behavior log information, obtain operation behaviors within a preset time interval after the user inputs the search keywords from the operation behavior log information, and combine the operations The operation object corresponding to the behavior is used as the initial search object;

A calculation unit for calculating the confidence score of the initial search object;

The establishment unit is configured to establish a mapping relationship list between the search object and the search keyword according to the confidence score of the initial search object.

In a possible implementation manner of the second aspect, the calculation unit is specifically configured to:

Acquiring at least one behavior characteristic value of the initial search object from the operation behavior log information;

Calculate the confidence score of the initial search object according to the at least one behavior feature value and a preset weight coefficient corresponding to each behavior feature value.

In a possible implementation manner of the second aspect, the behavior characteristic value includes the use times of the initial search object, the average use time of the initial search object, and/or the proportion of the use times of the initial search object.

The confidence score of the initial search object is calculated according to the search keyword, the initial search object, and a preset prediction model, where the preset prediction model is the search keyword and the initial search keyword corresponding to the The search object and the confidence score are the training samples, which are obtained by training the learning model using machine learning or deep learning algorithms.

In a possible implementation manner of the second aspect, the establishing unit is specifically configured to:

A mapping relationship list between the search object and the search keyword is established according to the weight value of the search keyword.

In a possible implementation manner of the second aspect, the establishing unit is further configured to:

In a possible implementation manner of the second aspect, the determining module is specifically configured to:

The search object corresponding to the search keyword whose weight value is greater than the preset weight threshold is taken as the target search object.

In a possible implementation of the second aspect, the mapping relationship establishment module further includes a preprocessing unit for:

Correspondingly, the extraction unit is specifically used for:

The search keyword is obtained from the target search behavior log information, and the operation behavior within a preset time interval after the user inputs the search keyword is obtained from the target operation behavior log information.

In a possible implementation manner of the second aspect, the mapping relationship establishment module further includes a filtering unit, configured to:

Correspondingly, the extraction unit is specifically used for:

The operation object corresponding to the effective operation behavior is used as the initial search object.

In a third aspect, an embodiment of the present application provides a terminal device, including: a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor executes the computer program When realizing the search method of the first aspect mentioned above.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the search method as in the first aspect described above is implemented.

In the fifth aspect, embodiments of the present application provide a computer program product, which when the computer program product runs on a terminal device, causes the terminal device to execute the search method of the first aspect described above.

It is understandable that, for the beneficial effects of the second aspect to the fifth aspect described above, reference may be made to the relevant description in the first aspect described above, and details are not repeated here.

Description of the drawings

Fig. 1 is a structural block diagram of a terminal device provided by an embodiment of the present application;

FIG. 2 is a schematic flowchart of the search method provided by the first embodiment of the present application;

FIG. 3 is an application scenario diagram of the search method provided by an embodiment of the present application;

FIG. 4 is another application scenario diagram of the search method provided by the embodiment of the present application;

FIG. 5 is another application scenario diagram of the search method provided by the embodiment of the present application;

FIG. 6 is a schematic diagram of an operating environment of a search method provided by an embodiment of the present application;

Fig. 7 is a schematic flowchart of a method for establishing a mapping relationship list between search objects and search keywords provided by the first embodiment of the present application;

8 is a schematic flowchart of a method for establishing a mapping relationship list between search objects and search keywords provided by the second embodiment of the present application;

Fig. 9 is a schematic structural diagram of a search device provided by an embodiment of the present application.

Detailed ways

In the following description, for the purpose of illustration rather than limitation, specific details such as a specific system structure and technology are proposed for a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from obstructing the description of this application.

It should be understood that when used in the specification and appended claims of this application, the term "comprising" indicates the existence of the described features, wholes, steps, operations, elements and/or components, but does not exclude one or more other The existence or addition of features, wholes, steps, operations, elements, components, and/or collections thereof.

It should also be understood that the term "and/or" used in the specification and appended claims of this application refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

As used in the description of this application and the appended claims, the term "if" can be construed as "when" or "once" or "in response to determination" or "in response to detecting ". Similarly, the phrase "if determined" or "if detected [described condition or event]" can be interpreted as meaning "once determined" or "in response to determination" or "once detected [described condition or event]" depending on the context ]" or "in response to detection of [condition or event described]".

In addition, in the description of the specification of this application and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

The reference to "one embodiment" or "some embodiments" described in the specification of this application means that one or more embodiments of this application include a specific feature, structure, or characteristic described in combination with the embodiment. Therefore, the sentences "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. appearing in different places in this specification are not necessarily All refer to the same embodiment, but mean "one or more but not all embodiments" unless it is specifically emphasized otherwise. The terms "including", "including", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

The search method provided in the embodiments of this application can be applied to mobile phones, tablet computers, wearable devices, vehicle-mounted devices, augmented reality (AR)/virtual reality (VR) devices, notebook computers, and ultra-mobile personal computers ( Ultra-mobile personal computer (UMPC), netbook, personal digital assistant (personal digital assistant, PDA) and other terminal devices, can also be applied to smart home appliances such as speakers, TVs, washing machines, etc. The embodiments of this application are specific to terminal devices. There are no restrictions on the type.

Take the terminal device as a mobile phone as an example. Fig. 1 shows a block diagram of a part of the structure of a mobile phone provided by an embodiment of the present application. 1, the mobile phone includes: a radio frequency (RF) circuit 110, a memory 120, an input unit 130, a display unit 140, a sensor 150, an audio circuit 160, a wireless fidelity (WiFi) module 170, and a processor 180 , And power supply 190 and other components. Those skilled in the art can understand that the structure of the mobile phone shown in FIG. 1 does not constitute a limitation on the mobile phone, and may include more or fewer components than those shown in the figure, or a combination of some components, or different component arrangements.

The following describes the components of the mobile phone in detail with reference to Figure 1:

The RF circuit 110 can be used for receiving and sending signals during information transmission or communication. In particular, after receiving the downlink information of the base station, it is processed by the processor 180; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (LNA), a duplexer, and the like. In addition, the RF circuit 110 can also communicate with the network and other devices through wireless communication. The above-mentioned wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobile Communication (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (Code Division) Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), Email, Short Messaging Service (SMS), etc. For example, the user enters a contact on the interface of a call application program to search, the terminal device recommends a contact for the user according to the keyword of the contact entered by the user, the user dials the contact's phone, and the call is made through the RF circuit 110.

The memory 120 may be used to store software programs and modules. The processor 180 executes various functional applications and data processing of the mobile phone by running the software programs and modules stored in the memory 120. The memory 120 may mainly include a program storage area and a data storage area. The program storage area may store an operating system, an application program required by at least one function (such as a sound playback function, an image playback function, etc.), etc.; Data created by the use of mobile phones (such as audio data, phone book, etc.), etc. In addition, the memory 120 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

The input unit 130 may be used to receive inputted digital or character information, and generate key signal input related to user settings and function control of the mobile phone. Specifically, the input unit 130 may include a touch panel 131 and other input devices 132. The touch panel 131, also known as a touch screen, can collect user touch operations on or near it (for example, the user uses any suitable objects or accessories such as fingers, stylus, etc.) on the touch panel 131 or near the touch panel 131. Operation), and drive the corresponding connection device according to the preset program. Optionally, the touch panel 131 may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch position, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact coordinates, and then sends it To the processor 180, and can receive and execute the commands sent by the processor 180. In addition, the touch panel 131 can be implemented in multiple types such as resistive, capacitive, infrared, and surface acoustic wave. In addition to the touch panel 131, the input unit 130 may also include other input devices 132. Specifically, the other input device 132 may include, but is not limited to, one or more of a physical keyboard, function keys (such as volume control buttons, switch buttons, etc.), trackball, mouse, and joystick. For example, the user enters a search keyword through the input interface to search.

The display unit 140 may be used to display information input by the user or information provided to the user and various menus of the mobile phone. The display unit 140 may include a display panel 141. Optionally, the display panel 141 may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), etc. Further, the touch panel 131 can cover the display panel 141. When the touch panel 131 detects a touch operation on or near it, it transmits it to the processor 180 to determine the type of the touch event, and then the processor 180 responds to the touch event. The type provides corresponding visual output on the display panel 141. Although in FIG. 1, the touch panel 131 and the display panel 141 are used as two independent components to realize the input and input functions of the mobile phone, but in some embodiments, the touch panel 131 and the display panel 141 can be integrated. Realize the input and output functions of the mobile phone.

The mobile phone may also include at least one sensor 150, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor. The ambient light sensor can adjust the brightness of the display panel 141 according to the brightness of the ambient light. The proximity sensor can close the display panel 141 and/or when the mobile phone is moved to the ear. Or backlight. As a kind of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in various directions (usually three-axis), and can detect the magnitude and direction of gravity when it is stationary. Games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, percussion), etc.; as for other sensors such as gyroscopes, barometers, hygrometers, thermometers, infrared sensors, etc., which can also be configured in mobile phones, I will not here Go into details.

The audio circuit 160, the speaker 161, and the microphone 162 can provide an audio interface between the user and the mobile phone. The audio circuit 160 can transmit the electrical signal converted from the received audio data to the speaker 161, which is converted into a sound signal for output by the speaker 161; on the other hand, the microphone 162 converts the collected sound signal into an electrical signal, which is then output by the audio circuit 160. After being received, it is converted into audio data, and then processed by the audio data output processor 180, and then sent to, for example, another mobile phone via the RF circuit 110, or the audio data is output to the memory 120 for further processing. In the embodiment of the present application, the microphone 162 converts the collected sound signals into electrical signals, which are received by the audio circuit 160 and converted into audio data, and then the audio data is output to the processor 180, and the processor 180 performs corresponding searches based on the audio data.

WiFi is a short-distance wireless transmission technology. The mobile phone can help users send and receive emails, browse web pages, and access streaming media through the WiFi module 170. It provides users with wireless broadband Internet access. Although FIG. 1 shows the WiFi module 170, it is understandable that it is not a necessary component of a mobile phone, and can be omitted as needed without changing the essence of the invention.

The processor 180 is the control center of the mobile phone. It uses various interfaces and lines to connect various parts of the entire mobile phone. It executes by running or executing software programs and/or modules stored in the memory 120 and calling data stored in the memory 120. Various functions and processing data of the mobile phone can be used to monitor the mobile phone as a whole. Optionally, the processor 180 may include one or more processing units; preferably, the processor 180 may integrate an application processor and a modem processor, where the application processor mainly processes the operating system, user interface, application programs, etc. , The modem processor mainly deals with wireless communication. It can be understood that the foregoing modem processor may not be integrated into the processor 180.

The mobile phone also includes a power source 190 (such as a battery) for supplying power to various components. Preferably, the power source can be logically connected to the processor 180 through a power management system, so that functions such as charging, discharging, and power management can be managed through the power management system.

Although not shown, the mobile phone may also include a camera. Optionally, the position of the camera on the mobile phone may be front-mounted or rear-mounted, which is not limited in the embodiment of the present application. For example, a mobile phone scans a two-dimensional code through a camera, and performs a corresponding search based on the scanned two-dimensional code.

Optionally, the mobile phone may include a single camera, a dual camera, or a triple camera, etc., which is not limited in the embodiment of the present application.

For example, a mobile phone may include three cameras, of which one is a main camera, one is a wide-angle camera, and one is a telephoto camera.

Optionally, when the mobile phone includes multiple cameras, the multiple cameras may be all front-mounted, or all rear-mounted, or partly front-mounted and another part rear-mounted, which is not limited in the embodiment of the present application.

In addition, although not shown, the mobile phone may also include a Bluetooth module, etc., which will not be repeated here.

The search method provided by the embodiment of the present application will be described below with reference to FIG. 1. As shown in Figure 2, the search method provided by the embodiment of the present application includes:

S101: Obtain search keywords.

In an application scenario, when a user needs to search, he enters a voice, enters a character string, or scans a two-dimensional code on the search interface of the terminal device, and the terminal device extracts search keywords from the user input information.

In another application scenario, when the terminal device needs to recommend information for the user, such as recommending videos, products, services, etc., the user’s preferences can be obtained according to the user’s historical operation behavior, such as shopping records, video viewing records, etc. Like to get search keywords.

Exemplarily, as shown in FIG. 3, in an application scenario, the user enters "I want to take a taxi" on the application search interface to perform an application search, and the extracted search keyword is "take a taxi". As shown in Figure 4, in another application scenario, the user enters the voice "payment" through the voice assistant on the terminal device, and the terminal device recognizes the "payment" from the input voice and displays it in the search bar of the application. The search keyword for is "payment". As shown in Fig. 5, in another application scenario, the user enters "I want to see strong bald head" in the video search interface, and the search keyword extracted by the video application is "bald head strong".

S102: Determine the target search object corresponding to the search keyword according to the pre-established mapping relationship list between the search object and the search keyword; wherein, the mapping relationship list between the search object and the search keyword is based on historical search behaviors and search results. The operation behavior after the historical search behavior is established.

Among them, the search object can be applications, music, videos, news, web pages, and so on. The list of mapping relations between search objects and search keywords may be established by the terminal device counting the historical search behavior of the current user and the operation behavior after the historical search behavior; it may also be the server counting the historical search behavior and the history of all users. The operation behavior after the search behavior is established, and the established mapping relationship list is sent to the terminal device; it can also be that the server collects the historical search behaviors of all users and the operation behaviors after the historical search behavior, and the historical search behaviors and operations The behavior is sent to the terminal device, which is established by the terminal device's statistics of historical search behavior and operation behavior.

For example, as shown in FIG. 6, the terminal device 1 is in communication connection with the server 2, where the server 2 may be a server, or a server cluster composed of several servers, or a cloud computing service center. The server counts the historical search behaviors and operation behaviors of all terminal devices that are in communication with the server, and establishes a list of mapping relationships between search objects and search keywords. When the terminal device 1 obtains the search keyword, it obtains the mapping relationship list between the search object and the search keyword sent by the server. Among them, in the mapping relationship list between search objects and search keywords, one search keyword can correspond to multiple search objects, and the terminal device can regard all the multiple search objects corresponding to the search keywords as the target search objects, or combine the search objects with the search keywords. Part of the search object or one search object corresponding to the keyword is used as the target search object.

S103: Display the target search object.

Exemplarily, as shown in Fig. 3, in the mapping relationship list between search objects and search keywords, the target search objects corresponding to "Taxi" are "Didi Dache", "Quaigou Dache", "High German Map" and "Didi Dache". "Shenzhou Special Vehicle", the terminal device displays the above application list according to the mapping relationship list between the search object and the search keyword. As shown in Figure 4, in the mapping relationship list between search objects and search keywords, the target search objects corresponding to "Pay" are "Alipay", "Wallet", "WeChat" and "Cloud QuickPass". The list of mapping relationships with search keywords displays the above application list. As shown in Figure 5, in the mapping relationship list between search objects and search keywords, the target search objects corresponding to "Bald Head Strong" are ""Bear-Haunted Adventure Diary 2", ""Bear-Haunted Adventure Diary", " In "The Complete Works of Bears" and "The Fantasy Space of Bears", the terminal device displays the above-mentioned video list according to the list of the mapping relationship between the search object and the search keyword.

In the foregoing embodiment, since the mapping relationship list between search objects and search keywords is established based on historical search behaviors and operation behaviors after the historical search behaviors, the search objects can reflect the real intention of the user after inputting the search keywords. When the user inputs a search keyword, the target search object corresponding to the search keyword is determined according to the mapping relationship list between the search object and the search keyword, which can recommend more accurate search objects for the user.

In the search method provided in the embodiments of the present application, the method for establishing a mapping relationship list between search objects and search keywords will be described in detail below.

As shown in FIG. 7, the method for establishing a mapping relationship list between search objects and search keywords provided by the first embodiment of this application includes:

S201: Acquire search behavior log information and operation behavior log information of the user.

Specifically, the search behavior log information stores the user's historical search behavior, and the search behavior log information includes any one or more of the user ID, search time, original search string, search exposure list, search click list, and the like. Wherein, the search exposure list stores the search objects recommended by the terminal device for the user according to the extracted search keywords, for example, a list of application programs recommended for the user according to the search keywords. The search click list stores: the search object clicked by the user from the search exposure list. When the user does not select the search object from the search exposure list, the record of the corresponding search click list is empty.

The operation behavior log information stores the user's historical operation behavior. The operation behavior log information includes one of the user ID, the time when the operation behavior occurred, the name of the search object corresponding to the operation behavior, the category of the search object, the tag of the search object, etc. Or multiple. Among them, the category of the search object and the label of the search object are generated according to the characteristics of the search object. For example, when the search object is the application "WeChat", the category of the search object is chat, and the tags of the search object are social, communication, scan, voice, etc.

In a possible implementation manner, in order to improve the accuracy of the mapping relationship list, search behavior log information and operation behavior log information of all terminal devices connected to the server can be acquired.

Table 1

For example, in a scenario where a user searches for an application program, a user enters the original search string "I want to take a taxi" on the application search interface of a terminal device. The recommended applications on the terminal device are Kuaigou taxi driver version, Kuaigou taxi, Kuaigou Taxi (Quick App), the application recommended on the terminal device forms a search exposure list, the user does not click the application from the exposure list, and the corresponding click list is empty. Record the time when the search is started, search the corresponding user ID, search the exposure list, search the click list, and get the first record in Table 1. In Table 1, each search record of each user ID is recorded in turn, and the search behavior log information of the application program as shown in Table 1 is obtained.

Correspondingly, the time that each terminal device uses the application, the name of the application, the category of the application, the label of the application, etc. are recorded, and the behavior log information of using the application shown in Table 2 is obtained.

Table 2

S202: Preprocessing the search behavior log information and the operation behavior log information to obtain target search behavior log information and target operation behavior log information.

Specifically, data cleaning or feature engineering processing is performed on the acquired search behavior log information and operation behavior log information, including removing error data, duplicate data, and abnormal data, etc., to obtain target search behavior log information and target operation behavior log information. For example, remove data records with incomplete fields, records with contradictory time records, and records with missing fields.

S203: Extract search keywords from the target search behavior log information, obtain operation behaviors within a preset time interval after the user enters the search keywords from the target operation behavior log information, and use the operation object corresponding to the operation behavior as Initial search object.

Specifically, the search time is extracted from the target search behavior log information, and the search keywords are extracted from the original search string input by the user. For example, the keyword extracted from "I want to take a taxi" is "Taxi", the keyword extracted from "Order for Me" is "Order", and the keyword extracted from "What's news today" is "News", etc. . At the same time, extract the time of the operation behavior, the duration of the operation behavior and the name of the operation object from the target operation behavior log information. For example, for the behavior of searching for applications, the name of the application opened by the user and the time when the application was started to be used are extracted from the log information of the target operation behavior. According to the search time and the occurrence time of the operation behavior, the operation objects corresponding to the operation behavior after the user input the original search string are counted each time, and the operation objects corresponding to all the operation behaviors within the preset time interval after the original search string is input As the initial search object.

Exemplarily, for the operation behavior of using the application program, the application program used within a preset time interval after the user inputs the original search string is counted, and the calculated application program is used as the application program corresponding to the search keyword. For example, if the time interval is set to 30 seconds, the original search string entered by a user is "I want to take a taxi", the search time is "20190620 15:00:00", and the extracted search keyword is "take a taxi", then Count the applications used in the time period "20190620 15:00:00 to 20190620 15:00:30" on the terminal device, and the applications in this time period are the initial search objects.

In a possible implementation, in order to improve the accuracy of the calculation results, all terminal devices connected to the server are counted, and each search keyword corresponds to the initial search object, and the same search keyword of all users corresponds to the same The initial search objects of are merged, so that each search keyword corresponds to multiple initial search objects. For example, for the operation behavior of using the application, if the search keyword extracted from the user's original search string is "Taxi", the application used by the user within 30 seconds after the search is recorded; the search keyword for all users is For "Taxi", the applications used within 30 seconds after the search are counted, and the same applications are merged, and the statistical results shown in Table 3 are obtained.

table 3

Because some frequently used high-frequency operation behaviors often appear in the search list and interfere with the results; there are also some low-frequency operation behaviors that use less frequently, because the amount of data is too small, and the calculation error is large. In a possible implementation manner, after obtaining the operation behaviors within a preset time interval after the user enters the search keyword from the target operation behavior log information, delete the high-frequency operation behaviors whose operation frequency is greater than the first threshold and the delete operation frequency Low-frequency operation behaviors that are less than the second threshold obtain effective operation behaviors, and the operation object corresponding to the effective operation behavior is used as the initial search object, thereby improving the accuracy of calculation. Exemplarily, in the statistics of the application programs corresponding to the search keywords, the high-frequency application programs that are used frequently by the user and the low-frequency application programs that are used less frequently are removed. For example, the top 5 applications corresponding to "Taxi" in Table 3 are "WeChat", "Didi Travel", "Alipay", "High German Map", and "Today's Toutiao", among which "WeChat" and "Toutiao" "Alipay" and "Today's Toutiao" are high-frequency applications. When it is necessary to remove high-frequency operation behaviors with high operating frequency, these three applications are removed from the application list; when it is necessary to remove low-frequency operations with low operating frequency During the behavior, for example, you can remove applications that are used less than 1000 times, that is, applications that are used less frequently are removed from the list.

It can be understood that in another possible implementation manner, it is also possible to count only one initial search object corresponding to each search keyword. For example, for an application search with a search keyword of "Taxi", you can only count the application that was used first after the user entered the original search string, or the application that was used the most after the user entered the original search string, or the user input The application that has been used the longest after the original search string.

S204: Acquire at least one behavior characteristic value of the initial search object.

In a possible implementation manner, the behavior characteristic value is obtained after statistics of the initial search objects corresponding to the search keywords of all users. The behavior characteristic value includes the use times of the initial search object, the average use time of the initial search object, whether the initial search object is selected when displayed in the search exposure list, and/or the proportion of the use times of the initial search object. For example, for the initial search object "High German Map", the behavior feature values include the number of times the "High German Map" is used, the average duration of using the "High German Map", and whether there is "in the recommended list" when the search keyword is "Taxi". "High German Map", whether users clicked on "High German Map" from the recommended list, the percentage of usage of "High German Map", etc.

Exemplarily, after merging the initial search objects corresponding to the search keywords of all users, for a certain initial search object, calculate the average use time according to the operation time of all users on the initial search object; count all users operating the initial search object The number of searches for the object; calculate the proportion of the number of people operating the initial search object in the total number of people, and get the proportion of the number of times the initial search object is used. After all the records are merged, the user behavior wide table is obtained. For example, as shown in Table 4, in the application search, the search keyword is "Taxi". When the search keyword is "Taxi", count the applications used by all users within 30 seconds after the search, and calculate Corresponding to the "average duration of use", "number of times of use" and "proportion of times of use" of the corresponding application, the application width table corresponding to the search keyword "Taxi" is obtained.

Table 4

S205: Calculate the confidence score of the initial search object according to the at least one behavior feature value and a preset weight coefficient corresponding to each behavior feature value.

Specifically, the product of each behavior feature value and the corresponding preset weight coefficient is summed to obtain the confidence score of the corresponding initial search object.

For example, suppose that the behavioral characteristic value for calculating the confidence score of the initial search object includes the use times of the initial search object, the average use time of the initial search object, and the proportion of the use times of the initial search object. A certain search keyword corresponds to an initial search. The weighting coefficients of the number of use of the initial search object, the average use time of the initial search object, and the percentage of use of the initial search object are 0.4, 0.3, and 0.3, respectively. After normalizing each behavior feature value, the initial The use times of the search object, the average use time of the initial search object, and the use times of the initial search object account for 0.5, 0.5, and 0.1 respectively. Add the product of the value corresponding to each behavior characteristic value and the weight coefficient, that is, 0.4* 0.5+0.3*0.5+0.3*0.1=0.38, the confidence score is obtained.

For another example, in an application scenario, the behavior feature value for calculating the confidence score includes the proportion of the use times of the initial search object, and the confidence score is calculated according to the proportion of the use times of the initial search object. For example, in Table 5, In the statistics of the usage records of the application, 53% of the users who searched for "Taxi" used "Didi Travel", and the usage of "Didi Travel" corresponding to "Taxi" accounted for 0.53, then The confidence level of "Didi Travel" corresponding to "Taxi" is 0.53. In the same way, it is calculated that the usage frequency of the "High German Map" corresponding to "Taxi" is 0.25, and the confidence level of the "High German Map" corresponding to "Taxi" is 0.25, and each search keyword corresponding to each The confidence score of each application.

table 5

S206: According to the confidence score of the initial search object, establish a mapping relationship list between the search object and the search keyword.

In a possible implementation, the confidence score of the initial search object is converted into the confidence score of the search keyword, and the confidence score of the search keyword that is greater than the preset confidence threshold is used as the weight value of the search keyword . For example, the confidence score of the initial search object "Alipay" corresponding to the search keyword "Pay" is 0.35, and the confidence score of "Pay" in the search keyword corresponding to "Alipay" is 0.35. In addition, "Alipay" The corresponding search keywords include "ordering", "takeaway" and so on. The confidence scores of the initial search objects corresponding to the search keywords are sequentially converted into the confidence scores of the search keywords corresponding to the initial search objects. For each initial search object, sort the search keywords in descending order according to the corresponding confidence score, and use the confidence score in the record with the confidence score greater than the preset confidence threshold as the weight value, and each weight value corresponds to a record , Get a list of mapping relations between search objects and search keywords. That is, in the mapping relationship list between search objects and search keywords, each initial search object corresponds to multiple search keywords, and each search keyword corresponds to a weight value.

Table 6

For example, for application search, the confidence score of the application in Table 5 is converted into the confidence score of the search keyword, and after the confidence score is sorted in descending order, the confidence score in the top 5 records is the search The weight value of the keywords. Among them, the multiple search keywords corresponding to "DiDi Travel" and the corresponding weight values are ("DiDi", 0.9), ("Travel", 0.56), ("Taxi", 0.53 ), ("Driving", 0.45), ("Takeaway", 0.25), the search keyword is the intent tag. The weight value of each search keyword corresponding to each application is calculated, and the mapping relationship list between the application and the search keyword shown in Table 6 is obtained.

When the user conducts a search next time, the search keyword is extracted from the original search string, and the search corresponding to the record whose weight value of the search keyword is greater than the preset weight threshold according to the mapping relationship list between the search object and the search keyword The object is the target search object and recommended to the user. For example, when the user enters "Taxi" to search for applications, the applications corresponding to "Taxi" are sorted according to the weight value, and the applications corresponding to the top 5 records after sorting are recommended to the user.

In the above embodiment, by acquiring the user's search behavior log information and operation behavior log information, the operation behavior within a preset time interval after the search keyword is input is extracted from the log information, and the operation object corresponding to the operation behavior is taken as For the initial search object, the confidence score of the initial search object is calculated according to at least one behavior characteristic value of the initial search object and a preset weight coefficient corresponding to each behavior characteristic value. Since the behavior characteristic value represents the characteristic information of the initial search object, the confidence score calculated according to the behavior characteristic value represents the user's intention to operate the initial search object. According to the confidence score of the initial search object, the mapping relationship list between the search object and the search keyword is established, which can accurately reflect the user's real search intention. When the user searches, the list of the mapping relationship between the search object and the search keyword can be the user Recommend more accurate services.

As shown in FIG. 8, the search method provided by the second embodiment of the present application includes:

S301: Acquire search behavior log information and operation behavior log information of the user.

S302: Preprocessing the search behavior log information and the operation behavior log information to obtain the target search behavior log information and the target operation behavior log information.

S303: Extract search keywords from the target search behavior log information, obtain operation behaviors within a preset time interval after the user enters the search keywords from the target operation behavior log information, and use the operation object corresponding to the operation behavior as Initial search object.

Among them, S301-S303 are the same as S201-S203, and will not be repeated here.

S304: Calculate the confidence score of the initial search object according to the search keyword, the initial search object, and a preset prediction model.

Wherein, the preset prediction model is obtained by using search keywords, initial search objects corresponding to the search keywords, and confidence scores as training samples, and training the learning model using machine learning or deep learning algorithms. Specifically, the search keyword and the initial search object corresponding to the search keyword are input to the learning model, the characteristics of the search keyword, the characteristics of the initial search object, and the characteristics of the operation behavior corresponding to the search keyword are extracted, and the corresponding confidence score is output Optimize the parameters of the learning model according to the difference between the output confidence score and the confidence score in the training sample to obtain the optimal parameters of the learning model, and generate the preset prediction model according to the optimal parameters. Among them, the search keyword feature is any one or more of the popularity of the keyword, the total number of searches, the proportion of the number of searches, the word2vec word vector corresponding to the search keyword, and the characteristics of similar keywords of the search keyword. The characteristics of the initial search object include any one or more of the name, category, label of the search object, the corresponding word2vec word vector, and the characteristics of the similar search object of the search object. The operating behavior is characterized by any one or more of the average use time of the initial search object after the search, the popularity of use, the total number of uses, and the proportion of the number of uses.

In one possible implementation, the learning model can be a machine learning model such as logistic regression, gradient boosting tree, random forest, etc., or it can be a convolutional neural network model (Convolutional Neural Network, CNN), a fully connected neural network model (Fully For deep learning models such as Connected Neural Network (FCNN), the method of training the learning model can be a supervised learning algorithm or a semi-supervised learning algorithm. Exemplarily, a semi-supervised classification algorithm is used to train the learning model. First, for a first preset number of initial search objects, according to each search keyword corresponding to the initial search object, set each search keyword corresponding to each search keyword. The confidence score of the initial search object. For example, the number of applications for which the confidence score is to be calculated is 1,000, and 100 applications are selected from them. The confidence level of "Didi Travel" corresponding to "Taxi" is set to 0.5, and the confidence level of "Didi Travel" corresponding to "Taxi" is set to 0.5. The confidence score of “German Map” is 0.3, the confidence score of “Alipay” corresponding to “Pay” is 0.4, and the confidence score of “WeChat” corresponding to “Pay” is 0.4. In this way, 100 applications and The confidence score of the corresponding search keyword. The first preset number of initial search objects, search keywords, and confidence scores are used as training samples, and the learning model is trained and learned to obtain the first candidate model. Then, corresponding training samples are generated according to the second preset number of initial search objects, and the first candidate model is "re-learned" to optimize the parameters of the first candidate model and at the same time correct the inaccurate confidence score. For example, after selecting 100 applications to train the first candidate model from 1000 applications for which confidence scores are to be calculated, the remaining 900 applications are input into the first candidate model, according to the confidence level corresponding to each application Score, select the 100 applications with the highest credibility, and generate training samples again based on the 100 applications with the highest credibility and the 100 applications used to train the first candidate model for the first candidate model "Re-learning" to optimize the parameters of the first candidate model to obtain the second candidate model. This method is used to iteratively calculate in order to obtain the prediction model. Among them, the selection method of the application with the highest credibility may be selected based on the difference between the output results of multiple learning models. For example, the number of first candidate models is 3, and the remaining 900 applications are input to 3 first candidate models respectively, and the confidence score corresponding to the smallest difference between the output results of the 3 first candidate models is selected Application, as the most reliable application. Input the search keywords and the initial search object into the prediction model to calculate the confidence score of the initial search object.

S305: Establish a mapping relationship list between the search object and the search keyword according to the confidence score of the initial search object.

Among them, S305 is the same as S206, and will not be repeated here.

In the above embodiment, by acquiring the user's search behavior log information and operation behavior log information, the operation behavior within a preset time interval after the search keyword is input is extracted from the log information, and the operation object corresponding to the operation behavior is taken as For the initial search object, the confidence score of the initial search object is calculated according to the search keywords, the initial search object and the preset prediction model. Since the preset prediction model is generated by counting a large amount of data and repeatedly training, the calculation result is stable and universal , According to the confidence score of the initial search object, the mapping relationship list between the search object and the search keyword is established, which can accurately reflect the user's real search intent. When the user searches, the list of the mapping relationship between the search object and the search keyword can be Users recommend more accurate services.

It should be understood that the size of the sequence number of each step in the foregoing embodiment does not mean the order of execution. The execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

Corresponding to the search method described in the above embodiment, FIG. 9 shows a structural block diagram of a search device provided in an embodiment of the present application. For ease of description, only the parts related to the embodiment of the present application are shown.

Referring to Figure 9, the search device includes:

The obtaining module 10 is used to obtain search keywords;

The determining module 20 is configured to determine the target search object corresponding to the search keyword according to a pre-established list of mapping relations between search objects and search keywords; wherein, the list of mapping relations between search objects and search keywords is based on history The search behavior and the operation behavior after the historical search behavior is established;

The display module 30 is used to display the target search object.

In a possible implementation manner, the search device further includes a mapping relationship establishment module, and the mapping relationship establishment module includes:

In a possible implementation manner, the calculation unit is specifically configured to:

In a possible implementation manner, the behavior characteristic value includes the use times of the initial search object, the average use time of the initial search object, and/or the proportion of the use times of the initial search object.

In a possible implementation manner, the establishing unit is specifically configured to:

In a possible implementation manner, the establishing unit is further configured to:

In a possible implementation manner, the determining module is specifically configured to:

In a possible implementation manner, the mapping relationship establishment module further includes a preprocessing unit for:

Preprocessing the search behavior log information and the operation behavior log information respectively to obtain pre-processed target search behavior log information and target operation behavior log information;

Correspondingly, the extraction unit is specifically used for:

In a possible implementation manner, the mapping relationship establishment module further includes a filtering unit for:

Correspondingly, the extraction unit is specifically used for:

It should be noted that the information interaction and execution process between the above-mentioned devices/units are based on the same concept as the method embodiment of this application, and its specific functions and technical effects can be found in the method embodiment section. I won't repeat it here.

Those skilled in the art can clearly understand that, for the convenience and conciseness of description, only the division of the above functional units and modules is used as an example. In practical applications, the above functions can be allocated to different functional units and modules as needed. Module completion, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist alone physically, or two or more units can be integrated into one unit. The above-mentioned integrated units can be hardware-based Form realization can also be realized in the form of software functional unit. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the present application. For the specific working process of the units and modules in the foregoing system, reference may be made to the corresponding process in the foregoing method embodiment, which will not be repeated here.

In the above-mentioned embodiments, the description of each embodiment has its own focus. For parts that are not described in detail or recorded in an embodiment, reference may be made to related descriptions of other embodiments.

A person of ordinary skill in the art may realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.

In the embodiments provided in this application, it should be understood that the disclosed device/terminal device and method may be implemented in other ways. For example, the device/terminal device embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units. Or components can be combined or integrated into another system, or some features can be omitted or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.

In addition, the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.

If the integrated module/unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiments and methods, and can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the program is executed by the processor, it can implement the steps of the foregoing method embodiments. Wherein, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file, or some intermediate forms. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory) , Random Access Memory (RAM, Random Access Memory), electrical carrier signal, telecommunications signal, and software distribution media, etc.

The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, a person of ordinary skill in the art should understand that it can still implement the foregoing The technical solutions recorded in the examples are modified, or some of the technical features are equivalently replaced; these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the application, and should be included in Within the scope of protection of this application.

Claims

A search method, characterized in that it includes:

Get search keywords;

Determine the target search object corresponding to the search keyword according to a pre-established list of mapping relations between search objects and search keywords; wherein, the list of mapping relations between search objects and search keywords is based on historical search behaviors and the history Established by the operation behavior after the search behavior;

The target search object is displayed.
The search method according to claim 1, wherein the mapping relationship list between the search object and the search keyword is established in the following manner:

Obtain user search behavior log information and operation behavior log information;

The search keyword is obtained from the search behavior log information, the operation behavior within a preset time interval after the user inputs the search keyword is obtained from the operation behavior log information, and the operation object corresponding to the operation behavior is obtained As the initial search object;

Calculating the confidence score of the initial search object;

According to the confidence score of the initial search object, a mapping relationship list between the search object and the search keyword is established.
The search method according to claim 2, wherein the calculating the confidence score of the initial search object comprises:

Acquiring at least one behavior characteristic value of the initial search object from the operation behavior log information;

Calculate the confidence score of the initial search object according to the at least one behavior feature value and a preset weight coefficient corresponding to each behavior feature value.
The search method according to claim 3, wherein the behavior characteristic value includes the use times of the initial search object, the average use time of the initial search object, and/or the proportion of the use times of the initial search object.
The search method according to any one of claims 2 to 4, wherein the calculating the confidence score of the initial search object comprises:

The confidence score of the initial search object is calculated according to the search keyword, the initial search object, and a preset prediction model, where the preset prediction model is the search keyword and the initial search keyword corresponding to the The search object and the confidence score are the training samples, which are obtained by training the learning model using machine learning or deep learning algorithms.
3. The search method of claim 2, wherein the establishing a mapping relationship list between the search object and the search keyword according to the confidence score of the initial search object comprises:

Calculating the weight value of the search keyword according to the confidence score of the initial search object;

A mapping relationship list between the search object and the search keyword is established according to the weight value of the search keyword.
The search method according to claim 6, wherein the calculating the weight value of the search keyword according to the confidence score of the initial search object comprises:

Converting the confidence score of the initial search object into the confidence score of the search keyword;

The confidence score of the search keyword that is greater than the preset confidence threshold is used as the weight value of the search keyword.
The search method according to claim 7, wherein the determining the target search object corresponding to the search keyword comprises:

The search object corresponding to the search keyword whose weight value is greater than the preset weight threshold is taken as the target search object.
The search method according to any one of claims 2 to 8, characterized in that, before said obtaining search keywords from the search behavior log information, the method further comprises:

Preprocessing the search behavior log information and the operation behavior log information respectively to obtain preprocessed target search behavior log information and target operation behavior log information;

Correspondingly, obtaining search keywords from the search behavior log information, and obtaining operation behaviors within a preset time interval after the user inputs the search keywords from the operation behavior log information includes:

The search keyword is obtained from the target search behavior log information, and the operation behavior within a preset time interval after the user inputs the search keyword is obtained from the target operation behavior log information.
The search method according to any one of claims 2 to 9, characterized in that, after the operation behavior within a preset time interval after the user inputs the search keyword is obtained from the operation behavior log information, Also includes:

Delete operation behaviors whose operation frequency is greater than the first threshold and delete operation behaviors whose operation frequency is less than the second threshold from the acquired operation behaviors to obtain effective operation behaviors;

Correspondingly, using the operation object corresponding to the operation behavior as the initial search object includes:

The operation object corresponding to the effective operation behavior is used as the initial search object.
A search device, characterized in that it comprises:

Obtaining module for obtaining search keywords;

The determining module is used to determine the target search object corresponding to the search keyword according to a pre-established list of mapping relations between search objects and search keywords; wherein, the list of mapping relations between search objects and search keywords is based on historical search Behaviors and operational behaviors after the historical search behavior;

The display module is used to display the target search object.
A terminal device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program as claimed in claims 1 to 10. The method of any one of.
A computer-readable storage medium storing a computer program, wherein the computer program implements the method according to any one of claims 1 to 10 when the computer program is executed by a processor.