CN111782061A

CN111782061A - Method and device for recommending input mode of smart watch

Info

Publication number: CN111782061A
Application number: CN202010615171.0A
Authority: CN
Inventors: 王璟铭; 葛翔; 陈宪涛; 徐濛
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2020-06-30
Filing date: 2020-06-30
Publication date: 2020-10-16
Anticipated expiration: 2040-06-30
Also published as: CN111782061B

Abstract

The application discloses a method and a device for recommending an input mode of a smart watch, electronic equipment and a storage medium, and relates to the technical field of input methods. The specific implementation scheme is as follows: acquiring watch state information; determining the scene of the user according to the watch state information; in response to receiving the message, determining a message type of the message; acquiring a pre-trained historical preference model; and inputting the message type and the scene into a historical preference model, and determining a recommended input mode and/or contents of quick response. The implementation mode intelligently judges the most suitable input mode or the reply content at the moment, thereby avoiding the manual switching of different input modes by a user.

Description

Method and device for recommending input mode of smart watch

Technical Field

The embodiment of the application relates to the technical field of computers, in particular to the technical field of input methods.

Background

The smart watch serves as auxiliary wearable equipment of the smart phone, provides functions of message receiving, reminding and replying, and meets the demand of instant messaging when a user inconveniently uses the smart phone. However, due to the limitation of the screen size of the smart watch, the conventional keyboard/handwriting input has low input efficiency on the smart watch, and the voice input or quick reply with higher input efficiency cannot meet the requirements of users in many input scenes due to the situation limitation. Therefore, the user often needs to switch between different input modes to match the input mode according with the current use situation and personal habits.

The input mode that present intelligent wrist-watch supported mainly has: 1) the intelligent watch comprises a voice-based input mode, a voice message input mode and a voice-to-text input mode, wherein the voice-based input mode comprises a voice message or a voice-to-text input mode, the user inputs the message content in the voice mode, and the intelligent watch converts the voice into text and then sends the text; 2) quick response, wherein the watch provides a plurality of response contents defined in advance for the user to select; 3) keyboard input, which is to adapt a full keyboard or a Sudoku keyboard for pinyin input of the mobile phone to a watch interface; 4) and (4) handwriting input, namely performing handwriting input on the watch dial.

When a user needs to input, the input mode of the smart watch is presented based on two fixed strategies: 1) providing all input modes, and enabling a user to manually select before input is needed each time; 2) and providing a default input mode, entering the input mode interface by default when input is needed, and reserving an entrance for switching to other input modes.

Disclosure of Invention

The present disclosure provides a method, apparatus, device, and storage medium for recommending an input mode for a smart watch.

According to a first aspect of the present disclosure, there is provided a method for recommending an input mode of a smart watch, comprising: acquiring watch state information; determining the scene of the user according to the watch state information; in response to receiving the message, determining a message type of the message; acquiring a pre-trained historical preference model; and inputting the message type and the scene into a historical preference model, and determining a recommended input mode and/or contents of quick response.

According to a second aspect of the present disclosure, there is provided an apparatus for recommending an input method of a smart watch, comprising: a state acquisition unit configured to acquire watch state information; the scene determining unit is configured to determine a scene where the user is located according to the watch state information; a type determination unit configured to determine a message type of a message in response to receiving the message; a model acquisition unit configured to acquire a pre-trained historical preference model; and the recommending unit is configured to input the message type and the scene into a history preference model and determine a recommended input mode and/or contents of quick response.

According to a third aspect of the present disclosure, there is provided an electronic apparatus, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of the first aspect.

According to a fourth aspect of the present disclosure, there is provided a non-transitory computer readable storage medium having stored thereon computer instructions, wherein the computer instructions are for causing a computer to perform the method of any one of the first aspects.

This application is directed against the text input problem of intelligent wrist-watch, has provided the individualized scheme that presents of intelligent wrist-watch input of sight intelligence, and through the wrist-watch sensor, the user independently sets up the schedule, and the APP that the cell-phone is running judges the sight that the user was located this moment to combine message type and user to the historical preference of input mode, judge the input mode that this moment is fit for or reply the content, convenience of customers replies the message on the wrist-watch.

It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.

Drawings

The drawings are included to provide a better understanding of the present solution and are not intended to limit the present application. Wherein:

FIG. 1 is an exemplary system architecture diagram in which one embodiment of the present application may be applied;

FIG. 2 is a flow diagram of one embodiment of a method for recommending an input mode for a smart watch according to the present application;

FIG. 3 is a schematic diagram of one application scenario of a method for recommending an input mode for a smart watch according to the present application;

FIG. 4 is a flow diagram of yet another embodiment of a method for recommending an input modality for a smart watch according to the present application;

FIG. 5 is a schematic diagram illustrating an embodiment of an apparatus for recommending input modes for a smart watch according to the present application;

fig. 6 is a block diagram of an electronic device for implementing a method for recommending an input mode for a smart watch according to an embodiment of the present application.

Detailed Description

The following description of the exemplary embodiments of the present application, taken in conjunction with the accompanying drawings, includes various details of the embodiments of the application for the understanding of the same, which are to be considered exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

Fig. 1 illustrates an exemplary system architecture 100 to which the methods and apparatuses for recommending an input modality for a smart watch of embodiments of the present application may be applied.

As shown in fig. 1, system architecture 100 may include

smart watches

101, 102, network 103, database server 104, and server 105. Network 103 is the medium used to provide communication links between

smart watches

101, 102, database server 104, and server 105. Network 103 may include various connection types, such as wired, wireless communication links, and so forth.

Users

110, 120 may interact with server 105 over network 103 using

smart watches

101, 102, respectively, to receive or send messages, etc. Various client applications, such as model training applications, input methods, shopping applications, payment applications, web browsers, instant messenger applications, etc., may be installed on

smartwatches

101, 102. Instant messages, e.g., WeChat, spike, QQ, etc., may be communicated between

smartwatches

101, 102 through server 105 or other servers.

The

smart watches

101, 102 herein may comprise various sensors, such as an acceleration sensor, a heart rate sensor, a GPS, a sound sensor (e.g. a microphone).

Database server 104 may be a database server that provides various services. For example, a database server may have a sample set stored therein. The sample set contains a large number of samples. The sample may include a scene, a message type, and an input mode or content of the quick reply corresponding to the scene and the message type. In this way, after the

users

110 and 120 select the input mode or the content of the quick reply after receiving the message on the

smartwatch

101 and 102, the current scene, the message type, and the selection result are stored as a sample in the database server 104.

Server 105 may also be a server that provides various services, such as a backend server that provides support for historical preference models used on

smartwatches

101, 102. The backend server may train the initial model using samples in the sample set sent by the

smart watches

101, 102 and may send the training results (e.g., the generated historical preference model) to the

smart watches

101, 102. In this way, the

smartwatch

101, 102 may apply the generated historical preference model for automatic recommendation of input modes or quick replies.

Here, the database server 104 and the server 105 may be hardware or software. When they are hardware, they can be implemented as a distributed server cluster composed of a plurality of servers, or as a single server. When they are software, they may be implemented as multiple pieces of software or software modules (e.g., to provide distributed services), or as a single piece of software or software module. And is not particularly limited herein.

It should be noted that, the method for recommending the input mode of the smart watch provided in the embodiment of the present application is generally performed by the smart watch. Accordingly, a device for recommending an input mode of a smart watch is generally provided in the smart watch.

It is noted that database server 104 may not be provided in system architecture 100, as server 105 may perform the relevant functions of database server 104.

It should be understood that the number of smartwatches, networks, database servers, and servers in fig. 1 are merely illustrative. There may be any number of smartwatches, networks, database servers, and servers, as desired for implementation.

With continued reference to FIG. 2, a flow 200 of one embodiment of a method for generating a face detection model according to the present application is shown. The method for generating a face detection model may comprise the steps of:

step 201, obtaining watch state information.

In this embodiment, an executing entity (for example, the smart watch shown in fig. 1) of the method for recommending an input mode of the smart watch may obtain watch state information, where the watch state information includes at least one of the following: sensor information, schedule information, running APP information.

1) Acquiring sensor information through a watch sensor: the acceleration sensor can be used for detecting the movement of a user, the heart rate sensor can be used for detecting the heart rate of the user, and the GPS can be used for positioning the user and the movement track of the user. In addition, the microphone of the watch may be used to measure the noise level of the environment in which it is located, and determine whether voice input is appropriate.

2) Acquiring schedule information through a schedule set by a user: at present, the smart watch supports schedule connection with a mobile phone, and reminding is carried out before the schedule starts.

3) The mobile phone runs APP: the APP operated by the mobile phone can also be used for assisting in judging the current state of the user.

Step 202, determining the scene of the user according to the watch state information.

In this embodiment, the scene where the user is located is mainly determined based on the following input sources:

1) through the sensor information, it can be judged that the user is: firstly, the heart rate is in a relatively static state, the heart rate is within a quiet heart rate, the GPS and the acceleration sensor display that the user is relatively static, and the user can be in relatively static activities such as working, meeting, class giving, watching television and the like; and secondly, the moving state, namely the GPS displays that the user is moving, and the heart rate and acceleration sensors can judge whether the user is walking, riding or in other moving modes according to the moving speed. The motion state can judge whether the user is in the motion state according to the heart rate, and outdoor motion or indoor motion can be further judged by combining the heart rate change rule and the GPS.

2) Whether the user is possibly in a busy state such as meeting and movement can be judged through a schedule set by the user, and then information obtained by the watch sensor is used for verification. When the sensor information is not enough to judge the specific state of the user, the judgment is assisted by schedule information. For example, if it is determined by the sensor that the user is in a relatively stationary state, in which the user may be in a relatively stationary activity such as working, meeting, class, watching tv, etc., if the schedule information shows that the user is in a meeting now, it can be confirmed that the user is in the meeting.

3) Judging whether the user is possibly in a motion state, a do-not-disturb state and a navigation state through running APP information, and verifying through information acquired by a watch sensor, wherein if the user is in the motion state, a mobile phone possibly starts a corresponding motion APP to provide motion guidance or record; while in the meeting, the handset may be in a do-not-disturb mode; while on the road, it may be in a map navigation mode. For example, if the user is determined to be in a mobile state by a sensor, the user may be walking, riding, or otherwise moving. If the running APP is a self-driving navigation, then it can be confirmed that the user is in the car.

In response to receiving the message, step 203 determines a message type of the message.

In this embodiment, the user can customize the granularity of the message type, and according to the defined type, a personalized input scheme can be provided. For example, according to the message source, dividing the message into working communication and private communication, or setting a preferential or non-preferential input mode for a specific object; the reply content of the message can be divided into concrete or non-concrete, single time or multiple times. Different message types, the user may have different input mode preferences. For example, messages from the nail are classified as work class communications and messages from the QQ are classified as private communications. The working class communication and the private communication can also be determined according to the same APP grouping, for example, the messages from the co-workers in the WeChat can be divided into the working class communication, and the messages from the family can be divided into the private communication. The leader may be set with non-priority voice input and the family may be set with priority voice input.

Optionally, the intention of the received message may be identified by semantic recognition techniques to determine whether the message reply content is specific or non-specific, single or multiple times. For example, if the received message is "get-ack at 4 pm, then it can be recognized that the message only requires a single reply of" get-back ". And if the received message is "which play you intend to go tomorrow? We trade off at one glance ", we can recognize that this message requires multiple replies. The intimacy of the received message may also be identified to determine the type of reply message, e.g., if the received message is "husband, morning next to work" and the relationship between them is identified, the contents of the quick reply may be added to "wife" automatically over the original "good", and the reply contents become "good, wife". Affinity may also be determined from contact groupings to prevent misdirection by received messages. For example, although the received message is "husband, morning next to work" but the person who sent the message is not wife, it cannot automatically add "wife" to reply. Therefore, better user experience is obtained, and the user receiving the reply does not feel the partner application.

Step 204, obtaining a pre-trained historical preference model.

In the embodiment, a cold start mode is adopted, and the input mode of the user for the current state preference is gradually learned. The method comprises the steps of presenting a conventional input scheme for a user to select, learning the selection of the user to judge the preference of the user for an input mode in a certain specific state, and gradually replacing the presented scheme with a personalized input scheme. The historical preference model can be a simple frequency statistical table or a trained neural network. The input method supports self-editing contents of quick response. The scene each time the user receives a message and the content of the selected quick reply may be recorded.

Step 205, inputting the message type and the scene into the history preference model, and determining the recommended input mode and/or the content of the quick reply.

In this embodiment, in combination with the judgment of the context where the user is located and the learning of the historical input preference of the user in different message types and contexts, a specific input mode and/or a quick reply option of personalized content may be presented. If some messages are not suitable for quick reply, the contents of the quick reply are not provided, if the messages are suitable for the quick reply, the contents of the quick reply can be presented preferentially to be selected by the user, and if the user does not select the quick reply, the input mode is switched to the recommended input mode. If the user selects the quick reply, the input mode does not need to be switched.

1) Presenting an input mode: at present, the user's autonomous input modes mainly include keyboard input (full keyboard/Sudoku), handwriting input, and voice input. They can be broadly divided into two categories: silent input (keyboard input and handwriting input) and non-silent input (voice input), generally speaking, for a certain user, silent input and non-silent input are in a complementary relationship and have different adaptability in different situations, while different input modes in silent input are in an alternative relationship with each other, and users often have a single preferred input mode in silent input. Therefore, the voice input or the silent input mode preferred by the user can be presented according to whether the scene where the user is located is suitable for the voice input and the input tendency of the user in the scene. In addition, the quick reply provides a message reply mode with non-autonomous input.

2) Quickly replying the content: the quick recovery mainly plays two roles: for a message of the notification class, an acknowledgement is indicated; for messages that require a specific reply, the message is delayed for a specific reply time while indicating receipt of the message. According to the current situation of the user, the quick reply can show the specific reason for delaying the reply, so that the quick reply is more true, and different intimacy exists according to personal habits when different message types are faced. Compared with the automatic input mode, the priority of the quick reply is different according to the historical preference of the user.

With continued reference to fig. 3, fig. 3 is a schematic diagram of an application scenario of the method for recommending an input mode of a smart watch according to the present embodiment. In the application scenario of fig. 3, the smart watch of the user acquires watch state information in real time. Detecting the heart rate 110 of the user through a heart rate sensor, detecting that the user is in a motion state based on the heart rate, and displaying that the user is in regular heart rate acceleration and recovery according to the change rule of the heart rate; detecting that a user is in a building through a GPS; the noise detection display of the microphone can be used for voice input; the schedule set by the user shows that the user is in motion and the mobile phone is running the exercise APP course, and it is inferred that the user may be performing instrument training in the gym.

According to the selection display of the historical input mode of the user, when the user trains the apparatus in the gymnasium, the user tends to use the quick input representation to confirm or delay the reply to the working message, when the quick reply is not suitable, the user tends to reply by the input mode of the Sudoku, and the user never adopts the voice input mode to the working message; whereas in the case of private messages, users always tend to reply with a voice input.

When the user receives the message from the nail, the user judges that the message is a work type message, and preferentially presents quick reply content according to the preference, and according to the state of the user, politely confirms that 'good, receiving' or 'I is moving', and replies to you after about half an hour. When the quick reply can not meet the requirement of the user, the input mode of the Sudoku can be quickly switched to. When the user receives the private message, the voice input mode is presented for the user to input according to the preference.

The method provided by the embodiment of the application comprehensively utilizes a plurality of information sources available by the existing intelligent watch, including information such as a watch sensor, a schedule set by a user, a connected mobile phone state and the like, provides a method for judging the situation of the user based on the available information, and intelligently provides an input mode of user tendency and contents of quick response by combining the message type and the user history input mode preference, so that the input information becomes more intelligent, personalized and efficient. The method and the device optimize the input information distribution process experience of the intelligent watch to a certain extent, and widen the use scene of the intelligent watch.

With further reference to fig. 4, a flow 400 of yet another embodiment of a method for recommending an input modality for a smart watch is illustrated. The process 400 of the method for recommending an input mode for a smart watch includes the steps of:

step 401, determining a network structure of the initial neural network and initializing network parameters of the initial neural network.

In the present embodiment, the execution main body of

steps

401 and 404 may be the same as or different from the execution main body of

steps

201 and 205. If the network structure information is the same as the parameter value of the network parameter, the execution subject of step 401-404 may store the network structure information of the trained neural network and the parameter value of the network parameter locally after the neural network is obtained by training. If not, the execution main body in the

step

401 and 404 may send the network structure information of the trained neural network and the parameter value of the network parameter to the execution main body of the method in the

step

201 and 205 after the neural network is obtained by training.

In this embodiment, the execution subject of

step

401 and 404 may first determine the network structure of the initial neural network. For example, it is necessary to determine which layers the initial neural network includes, the connection order relationship between layers, and which neurons each layer includes, the weight (weight) and bias term (bias) corresponding to each neuron, the activation function of each layer, and so on.

It will be appreciated that, since the neural network may comprise various types of neural networks, the network structure that needs to be determined is also different for different types of neural networks.

Step 402, a training sample set is obtained.

In this embodiment, the training sample set may be obtained locally or remotely from other electronic devices connected to the executive body network. The training sample comprises a scene, a message type and an input mode selected by a user or contents of quick response. For example, indoor sports scenes, work messages, user selection of quick reply "good, received" or selection of 9 grid input mode. The training sample set includes the user's selection of various message types under various scenarios. The sample can come from the user himself, and if the sample size is not enough, the selection of other users can be referred to.

And 403, taking scenes and message types in the training samples in the training sample set as the input of the initial neural network, taking the input mode selected by the user or the content of the quick reply as the expected output of the initial neural network, and training the initial neural network by using a machine learning method.

In this embodiment, the difference between the obtained input method or the content of the quick reply and the expected output in the training sample is first calculated by using a preset loss function, for example, the difference between the obtained input method or the content of the quick reply and the input method or the content of the quick reply in the training sample can be calculated by using the L2 norm as the loss function. Then, the network parameters of the initial neural network may be adjusted based on the calculated difference, and the training may be ended in case that a preset training end condition is satisfied. For example, the preset training end condition may include, but is not limited to, at least one of the following: the training time exceeds the preset time; the training times exceed the preset times; the calculated difference is less than a preset difference threshold.

Here, various implementations may be employed to adjust network parameters of the initial neural network based on differences between the calculated input manner or content of the quick reply and the desired output. For example, a BP (Back Propagation) algorithm or an SGD (Stochastic Gradient Descent) algorithm may be used to adjust the network parameters of the initial neural network.

And step 404, determining the initial neural network obtained by training as a pre-trained historical preference model.

In this embodiment, the initial neural network trained in step 403 may be determined as a pre-trained historical preference model.

As can be seen from fig. 4, compared with the embodiment corresponding to fig. 2, the flow 400 of the method for recommending an input mode of a smart watch in the present embodiment represents a step of generating a history preference model. Therefore, the scheme described in the embodiment can generate a history preference model for each user, and improve the hit rate of input modes or quick reply content recommendation.

With further reference to fig. 5, as an implementation of the method shown in the above figures, the present application provides an embodiment of an apparatus for recommending an input mode of a smart watch, where the apparatus embodiment corresponds to the method embodiment shown in fig. 2, and the apparatus may be applied to various electronic devices.

As shown in fig. 5, the apparatus 500 for recommending an input mode of a smart watch according to the present embodiment includes: a state acquisition unit 501, a scene determination unit 502, a type determination unit 503, a model acquisition unit 504, and a recommendation unit 505. Wherein, the state obtaining unit 501 is configured to obtain watch state information; a scene determining unit 502 configured to determine a scene where the user is located according to the watch state information; a type determining unit 503 configured to determine a message type of the message in response to receiving the message; a model obtaining unit 504 configured to obtain a pre-trained historical preference model; and a recommending unit 505 configured to input the message type and the scene into a history preference model, and determine a recommended input mode and/or contents of the quick reply.

In this embodiment, the specific processes of the state acquiring unit 501, the scenario determining unit 502, the type determining unit 503, the model acquiring unit 504 and the recommending unit 505 of the apparatus 500 for recommending a smart watch may refer to step 201, step 202, step 203, step 204 and step 205 in the corresponding embodiment of fig. 2.

In some optional implementations of this embodiment, the watch state information includes at least one of: sensor information, schedule information, running APP information.

In some optional implementations of the present embodiment, the state acquisition unit 501 is further configured to acquire the sensor information by at least one of: detecting movement information of a user through an acceleration sensor; detecting a heart rate of a user through a heart rate sensor; positioning a user and a moving track thereof through a GPS; the ambient noise level is measured by a microphone.

In some optional implementations of this embodiment, the scene determination unit 502 is further configured to: determining the scene of the user according to the sensor information; and verifying the scene where the user is located through the schedule information.

In some optional implementations of this embodiment, the scene determination unit 502 is further configured to: determining the scene of the user according to the sensor information; and verifying the scene where the user is located through the running APP information.

In some optional implementations of this embodiment, the type determining unit 503 is further configured to: dividing the message into a working class or a private class according to the source of the message; or setting a priority or non-priority input mode for a specific object; or semantically understanding the message, and determining the type of the reply content of the message, wherein the type is divided into specific or non-specific, single time or multiple times.

In some optional implementations of this embodiment, the apparatus 500 further comprises a model training unit (not shown in the drawings) configured to: determining a network structure of an initial neural network and initializing network parameters of the initial neural network; acquiring a training sample set, wherein the training sample set comprises a scene, a message type and an input mode selected by a user or contents of quick response; taking scenes and message types in training samples in a training sample set as input of an initial neural network, taking an input mode selected by a user or contents quickly replied as expected output of the initial neural network, and training the initial neural network by using a machine learning device; and determining the initial neural network obtained by training as a pre-trained historical preference model.

According to an embodiment of the present application, an electronic device and a readable storage medium are also provided.

Fig. 6 is a block diagram of an electronic device for recommending an input method of a smart watch according to an embodiment of the present application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application that are described and/or claimed herein.

As shown in fig. 6, the electronic apparatus includes: one or more processors 601, memory 602, and interfaces for connecting the various components, including a high-speed interface and a low-speed interface. The various components are interconnected using different buses and may be mounted on a common motherboard or in other manners as desired. The processor may process instructions for execution within the electronic device, including instructions stored in or on the memory to display graphical information of a GUI on an external input/output apparatus (such as a display device coupled to the interface). In other embodiments, multiple processors and/or multiple buses may be used, along with multiple memories and multiple memories, as desired. Also, multiple electronic devices may be connected, with each device providing portions of the necessary operations (e.g., as a server array, a group of blade servers, or a multi-processor system). In fig. 6, one processor 601 is taken as an example.

The memory 602 is a non-transitory computer readable storage medium as provided herein. Wherein the memory stores instructions executable by at least one processor to cause the at least one processor to perform the method for recommending input modalities for a smart watch provided herein. The non-transitory computer readable storage medium of the present application stores computer instructions for causing a computer to perform the method for recommending an input modality for a smart watch provided by the present application.

The memory 602 is used as a non-transitory computer readable storage medium and may be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as program instructions/modules corresponding to the method for recommending an input mode of a smart watch in the embodiment of the present application (for example, the state acquisition unit 501, the scene determination unit 502, the type determination unit 503, the model acquisition unit 504, and the recommendation unit 505 shown in fig. 5). The processor 601 executes various functional applications of the server and data processing by running non-transitory software programs, instructions and modules stored in the memory 602, that is, implementing the method for recommending an input mode of the smart watch in the above method embodiment.

The memory 602 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of the electronic device for recommending an input manner of the smart watch, and the like. Further, the memory 602 may include high speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, memory 602 optionally includes memory located remotely from processor 601, which may be connected via a network to an electronic device for recommending input modalities for a smart watch. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The electronic device for the method of recommending an input mode of a smart watch may further include: an input device 603 and an output device 604. The processor 601, the memory 602, the input device 603 and the output device 604 may be connected by a bus or other means, and fig. 6 illustrates the connection by a bus as an example.

The input device 603 may receive input numeric or character information and generate key signal inputs related to user settings and function controls of the electronic device for recommending input modalities for the smart watch, such as a touch screen, keypad, mouse, track pad, touch pad, pointer stick, one or more mouse buttons, track ball, joystick, or other input device. The output devices 604 may include a display device, auxiliary lighting devices (e.g., LEDs), and tactile feedback devices (e.g., vibrating motors), among others. The display device may include, but is not limited to, a Liquid Crystal Display (LCD), a Light Emitting Diode (LED) display, and a plasma display. In some implementations, the display device can be a touch screen.

Various implementations of the systems and techniques described here can be realized in digital electronic circuitry, integrated circuitry, application specific ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

These computer programs (also known as programs, software applications, or code) include machine instructions for a programmable processor, and may be implemented using high-level procedural and/or object-oriented programming languages, and/or assembly/machine languages. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or device (e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and/or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and/or data to a programmable processor.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

According to the technical scheme of the embodiment of the application, the intelligent watch input personalized presentation scheme with the intelligent scenes is provided, the schedule is set by a user independently through a watch sensor, the APP which is running by a mobile phone judges the scenes where the user is located at the moment, and the information type and the historical preference of the user on the input mode are combined to judge the suitable input mode or the reply content at the moment.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present application may be executed in parallel, sequentially, or in different orders, and the present invention is not limited thereto as long as the desired results of the technical solutions disclosed in the present application can be achieved.

The above-described embodiments should not be construed as limiting the scope of the present application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for recommending an input mode for a smart watch, comprising:

acquiring watch state information;

determining the scene of the user according to the watch state information;

in response to receiving a message, determining a message type of the message;

acquiring a pre-trained historical preference model;

and inputting the message type and the scene into the historical preference model, and determining a recommended input mode and/or contents of quick response.

2. The method of claim 1, wherein the watch state information comprises at least one of: sensor information, schedule information, running APP information.

3. The method of claim 2, wherein the sensor information is obtained by at least one of:

detecting movement information of a user through an acceleration sensor;

detecting a heart rate of a user through a heart rate sensor;

positioning a user and a moving track thereof through a GPS;

the ambient noise level is measured by a microphone.

4. The method of claim 2, wherein the determining a context in which a user is located from the watch state information comprises:

determining the scene where the user is located according to the sensor information;

and verifying the scene where the user is located through the schedule information.

5. The method of claim 2, wherein the determining a context in which a user is located from the watch state information comprises:

and verifying the scene where the user is located through the running APP information.

6. The method of claim 1, wherein the determining a message type of the message comprises:

dividing the message into a working class or a private class according to the source of the message; or

Setting a priority or non-priority input mode for a specific object; or

And performing semantic understanding on the message, and determining the type of the message reply content, wherein the type is divided into concrete or non-concrete, and is single or multiple.

7. The method of any of claims 1-6, wherein the historical preference model is trained by:

determining a network structure of an initial neural network and initializing network parameters of the initial neural network;

acquiring a training sample set, wherein the training sample set comprises a scene, a message type and an input mode selected by a user or contents of quick response;

taking scenes and message types in training samples in the training sample set as the input of the initial neural network, taking the input mode selected by a user or the content of quick reply as the expected output of the initial neural network, and training the initial neural network by using a machine learning method;

and determining the initial neural network obtained by training as the pre-trained historical preference model.

8. An apparatus for recommending an input modality for a smart watch, comprising:

a state acquisition unit configured to acquire watch state information;

the scene determining unit is configured to determine a scene where a user is located according to the watch state information;

a type determination unit configured to determine a message type of a message in response to receiving the message;

a model acquisition unit configured to acquire a pre-trained historical preference model;

and the recommending unit is configured to input the message type and the scene into the historical preference model and determine a recommended input mode and/or contents of quick response.

9. The apparatus of claim 8, wherein the watch state information comprises at least one of: sensor information, schedule information, running APP information.

10. The apparatus of claim 9, wherein the status acquisition unit is further configured to acquire sensor information by at least one of:

detecting movement information of a user through an acceleration sensor;

detecting a heart rate of a user through a heart rate sensor;

positioning a user and a moving track thereof through a GPS;

the ambient noise level is measured by a microphone.

11. The apparatus of claim 9, wherein the scene determination unit is further configured to:

12. The apparatus of claim 9, wherein the scene determination unit is further configured to:

13. The apparatus of claim 8, wherein the type determination unit is further configured to:

Setting a priority or non-priority input mode for a specific object; or

14. The apparatus according to any one of claims 8-13, wherein the apparatus further comprises a model training unit configured to:

taking scenes and message types in training samples in the training sample set as the input of the initial neural network, taking the input mode selected by a user or the content of quick reply as the expected output of the initial neural network, and training the initial neural network by using a machine learning device;

15. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

16. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-7.