Information retrieval methods, devices, electronic equipment and storage media

Abstract

This disclosure relates to an information retrieval method, apparatus, electronic device, and storage medium, comprising: in response to an identification card information retrieval operation, acquiring feature information of the electronic device in multiple dimensions, wherein the electronic device stores multiple identification card information; inputting the feature information into a corresponding information retrieval model to obtain the current usage probability of each identification card information output by the information retrieval model; and retrieving the first target identification card information with the highest current usage probability among the multiple identification card information. By inputting the acquired feature information of the electronic device in multiple dimensions into the corresponding information retrieval model to obtain the current usage probability of each identification card information output by the information retrieval model, and retrieving the first target identification card information with the highest current usage probability among the multiple identification card information, the reliance on setting a retrieval time period and retrieval location range for identification card information retrieval can be avoided, and the accuracy of retrieving identification card information can be improved.

Description

Technical Field

[0001] This disclosure relates to the field of near-field communication technology, and in particular to information retrieval methods, apparatus, electronic devices, and storage media. Background Technology

[0002] NFC (Near Field Communication) cards are a product of NFC technology development and can be applied to mobile payments, electronic ticketing, access control, and other fields. In these scenarios, the information from the identification card can be read into the terminal device, eliminating the need for a physical identification card during use; the terminal device can directly replace the card, improving convenience. However, if the terminal device reads and stores information from multiple identification cards, the user needs to manually select the appropriate card.

[0003] In related technologies, a time period or location range for retrieving each identification card information is set. If the current time falls within the time period or the current location falls within the location range, the corresponding identification card information is retrieved. However, if the time period or location range is too close, incorrect identification card information may be retrieved, thus requiring the user to manually select the identification card information to be retrieved. Summary of the Invention

[0004] To overcome the problems existing in related technologies, this disclosure provides an information retrieval method, apparatus, electronic device, and storage medium.

[0005] According to a first aspect of the present disclosure, an information retrieval method is provided, comprising:

[0006] In response to an identification card information retrieval operation, feature information of the electronic device in multiple dimensions is obtained, wherein the electronic device stores multiple identification card information;

[0007] The feature information is input into the corresponding information retrieval model to obtain the current usage probability of each piece of identification card information output by the information retrieval model.

[0008] The first target identification card information with the highest current usage probability is retrieved from the multiple identification card information.

[0009] Optionally, the method further includes:

[0010] When the electronic device is successfully identified by the identifier, the sample feature information of the electronic device in the multiple dimensions and the sample identification card information of the successfully identified device are obtained.

[0011] The sample identification card information is used as the data label for the sample feature information to obtain the model training samples;

[0012] The information retrieval model is trained based on the training samples of the model.

[0013] Optionally, there are multiple information retrieval models, and each of the multiple information retrieval models corresponds one-to-one with the multiple identification card information. The step of inputting the feature information into the corresponding information retrieval model to obtain the current usage probability of each identification card information output by the information retrieval model includes:

[0014] The feature information in the multiple dimensions is input into each of the information retrieval models to obtain the current usage probability of the corresponding identification card information output by each of the information retrieval models.

[0015] Optionally, each of the information retrieval models includes a sub-model corresponding to each of the dimensions, and the sub-model can determine the normal distribution model of the sample feature information corresponding to the sub-model in the model training samples during the training process;

[0016] The step of inputting the feature information from the multiple dimensions into each of the information retrieval models to obtain the current usage probability of the corresponding identification card information output by each of the information retrieval models includes:

[0017] For any of the aforementioned information retrieval models, the feature information of each of the aforementioned dimensions is input into the corresponding sub-model in the information retrieval model to obtain the usage sub-probability output by each of the aforementioned sub-models;

[0018] Based on the usage sub-probability output by each sub-model, the current usage probability of the identification card information corresponding to the information call model is determined.

[0019] Optionally, the feature information in the multiple dimensions includes at least two of the following: the location information of the electronic device when responding to the identification card information call operation; the timestamp when responding to the identification card information call operation; the posture information of the electronic device when responding to the identification card information call operation; the motion state information of the user carrying the electronic device within a preset time period before responding to the identification card information call operation; and the up-and-down status information of the user carrying the electronic device within the preset time period before responding to the identification card information call operation.

[0020] Optionally, the method further includes:

[0021] If the verification of the first target identification card information fails, the user is prompted to manually switch the identification card information;

[0022] The manually switched identification card information is used as the second target identification card information.

[0023] Optionally, the method further includes:

[0024] Based on the feature information corresponding to the verification failure, the sample feature information in the information retrieval model corresponding to the first target identification card information is removed to update the model training samples in the information retrieval model corresponding to the first target identification card information, and the sample feature information in the information retrieval model corresponding to the second target identification card information is added to update the model training samples in the information retrieval model corresponding to the second target identification card information.

[0025] Optionally, the method further includes:

[0026] If the number of verification failures for the same identification card information exceeds a preset threshold, the information retrieval model corresponding to the identification card information is retrained based on the updated model training samples, and the adjustment parameters and observation parameters of the information retrieval model are updated.

[0027] The adjustment parameters are used to adjust the sub-model of the retrained information retrieval model to a standard normal distribution model, and the observation parameters are the weights of the retrained information retrieval model when calculating the current usage probability.

[0028] According to a second aspect of the present disclosure, an information retrieval device is provided, comprising:

[0029] The acquisition module is configured to acquire feature information of the electronic device in multiple dimensions in response to an identification card information call operation, wherein the electronic device stores multiple identification card information.

[0030] The input module is configured to input the feature information into the corresponding information calling model to obtain the current usage probability of each piece of identification card information output by the information calling model.

[0031] The calling module is configured to call the first target identification card information with the highest current usage probability among the multiple identification card information.

[0032] Optionally, the input module is further configured to:

[0033] When the electronic device is successfully identified by the identifier, the sample feature information of the electronic device in the multiple dimensions and the sample identification card information of the successfully identified device are obtained.

[0034] The sample identification card information is used as the data label for the sample feature information to obtain the model training samples;

[0035] The information retrieval model is trained based on the training samples of the model.

[0036] Optionally, there are multiple information retrieval models, and each of the multiple information retrieval models corresponds one-to-one with the multiple identification card information. The input module is configured to input the feature information in the multiple dimensions into each of the information retrieval models respectively, so as to obtain the current usage probability of the corresponding identification card information output by each of the information retrieval models.

[0037] Optionally, each of the information retrieval models includes a sub-model corresponding to each of the dimensions, and the sub-model can determine the normal distribution model of the sample feature information corresponding to the sub-model in the model training samples during the training process;

[0038] The input module is configured to input feature information in each dimension into the corresponding sub-model of the information retrieval model for any of the information retrieval models, and obtain the usage sub-probability output by each sub-model;

[0039] Based on the usage sub-probability output by each sub-model, the current usage probability of the identification card information corresponding to the information call model is determined.

[0040] Optionally, the feature information in the multiple dimensions includes at least two of the following: the location information of the electronic device when responding to the identification card information call operation; the timestamp when responding to the identification card information call operation; the posture information of the electronic device when responding to the identification card information call operation; the motion state information of the user carrying the electronic device within a preset time period before responding to the identification card information call operation; and the up-and-down status information of the user carrying the electronic device within the preset time period before responding to the identification card information call operation.

[0041] Optionally, the device further includes a prompting module configured to prompt the user to manually switch the identification card information if the verification of the first target identification card information fails;

[0042] The calling module is also configured to call the manually switched identification card information as the second target identification card information.

[0043] Optionally, the device further includes an update module, configured to remove sample feature information in the information retrieval model corresponding to the first target identification card information based on the feature information corresponding to the verification failure, so as to update the model training samples in the information retrieval model corresponding to the first target identification card information, and to add sample feature information in the information retrieval model corresponding to the second target identification card information, so as to update the model training samples in the information retrieval model corresponding to the second target identification card information.

[0044] Optionally, the update module is further configured to, when the number of verification failures for the same identification card information exceeds a preset threshold, retrain the information retrieval model corresponding to the identification card information based on the updated model training samples, and update the adjustment parameters and observation parameters of the information retrieval model;

[0045] The adjustment parameters are used to adjust the sub-model of the retrained information retrieval model to a standard normal distribution model, and the observation parameters are the weights of the retrained information retrieval model when calculating the current usage probability.

[0046] According to a third aspect of the present disclosure, an electronic device is provided, comprising:

[0047] processor;

[0048] Memory used to store processor-executable instructions;

[0049] The processor is configured as follows:

[0050] In response to an identification card information retrieval operation, feature information of the electronic device in multiple dimensions is obtained, wherein the electronic device stores multiple identification card information;

[0051] The feature information is input into the corresponding information retrieval model to obtain the current usage probability of each piece of identification card information output by the information retrieval model.

[0052] The first target identification card information with the highest current usage probability is retrieved from the multiple identification card information.

[0053] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided that stores computer program instructions thereon, which, when executed by a processor, implement the steps of the method described in any one of the first aspects.

[0054] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects:

[0055] By responding to identification card information retrieval operations, the system acquires feature information of the electronic device across multiple dimensions. The electronic device stores multiple identification card information entries. This feature information is input into the corresponding information retrieval model to obtain the current usage probability of each identification card entry. Finally, the system retrieves the first target identification card entry with the highest current usage probability from among the multiple entries. By inputting the acquired feature information of the electronic device across multiple dimensions into the corresponding information retrieval model to obtain the current usage probability of each identification card entry, and retrieving the first target identification card entry with the highest current usage probability from among the multiple entries, the system avoids relying on setting a retrieval time period and location range for identification card information retrieval, and also improves the accuracy of retrieving identification card information.

[0056] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0057] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0058] Figure 1 This is a flowchart illustrating an information retrieval method according to an exemplary embodiment.

[0059] Figure 2 This is a flowchart illustrating another information retrieval method according to an exemplary embodiment.

[0060] Figure 3 This is a flowchart illustrating another information retrieval method according to an exemplary embodiment.

[0061] Figure 4 This is a flowchart illustrating an information retrieval model training method according to an exemplary embodiment.

[0062] Figure 5 This is a flowchart illustrating another information retrieval method according to an exemplary embodiment.

[0063] Figure 6 This is a block diagram illustrating an information retrieval device according to an exemplary embodiment.

[0064] Figure 7 This is a block diagram illustrating an apparatus for information retrieval according to an exemplary embodiment. Detailed Implementation

[0065] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0066] Figure 1 This is a flowchart illustrating an information retrieval method according to an exemplary embodiment. This method can be used in a terminal, such as a smartwatch or smartphone. Figure 1 As shown, it includes the following steps.

[0067] In step S11, in response to the identification card information call operation, the feature information of the electronic device in multiple dimensions is obtained.

[0068] The electronic device stores multiple identification card information entries. Each identification card information entry can be entered manually or automatically. The identification card information can be NFC card information. For example, by selecting NFC card information entry on the terminal's interface, and upon NFC card recognition, the NFC card information is read and stored.

[0069] Optionally, the card information retrieval operation can be the recognition action of the terminal being recognized, such as the recognition action of the smartwatch when it is brought close to the recognizer, or it can be the user retrieving the card information in the terminal through a preset gesture, such as retrieving the card information by pressing the volume up button twice when the terminal screen is off.

[0070] Based on the above embodiments, the feature information in the multiple dimensions includes at least two of the following: the location information of the electronic device when responding to the identification card information call operation; the timestamp when responding to the identification card information call operation; the posture information of the electronic device when responding to the identification card information call operation; the motion state information of the user carrying the electronic device within a preset time period before responding to the identification card information call operation; and the up-and-down status information of the user carrying the electronic device within the preset time period before responding to the identification card information call operation.

[0071] The feature information in multiple dimensions can include feature information in two dimensions, three dimensions, four dimensions, or all dimensions.

[0072] With user authorization, the location information can be obtained via the electronic device's GPS (Global Positioning System) in response to a card information call operation.

[0073] The timestamp can be obtained through the clock built into the electronic device, or by obtaining world time as the timestamp when responding to the card information call operation.

[0074] Specifically, the attitude information of the electronic device in response to the card information call operation is determined by collecting attitude angle information from the gyroscope built into the electronic device. For example, the angle between the electronic device and the horizontal or vertical direction is determined, thereby determining the attitude information of the electronic device in response to the card information call operation.

[0075] Specifically, by using acceleration information collected by the accelerometer built into the electronic device and attitude angle information collected by the gyroscope, the motion state information of the user carrying the electronic device is determined within a preset time period before responding to the identification card information call operation. For example, the motion state information can include being stationary, walking, running, cycling, going up and down stairs, driving, taking public transportation, taking the subway, etc.

[0076] Specifically, the user's status information regarding going up and down stairs within a preset time period prior to responding to the identification card information retrieval operation can be determined by using the air pressure value of the environment where the electronic device is located, collected by a barometer. For example, the user's status information regarding going up and down stairs within a preset time period prior to responding to the identification card information retrieval operation can be determined by the positive or negative relationship of the air pressure difference collected successively within the preset time period obtained by the barometer.

[0077] For example, if the difference between the air pressure value obtained at an earlier time and the air pressure value obtained at a later time is greater than 0 within a preset time period, the user's status information for going up or down stairs within the preset time period before responding to the card information call operation is determined to be going upstairs. If the difference between the air pressure value obtained at an earlier time and the air pressure value obtained at a later time is less than 0, the user's status information for going up or down stairs within the preset time period before responding to the card information call operation is determined to be going downstairs.

[0078] It can be explained that the status information of going up or down the stairs can narrow down the scope of card information retrieval. For example, when the user is going upstairs, it is usually after getting off the subway that the subway card needs to be retrieved, or after entering the office building that the company access card needs to be retrieved. When the user is going downstairs, it is usually before leaving the company that the office building access card needs to be retrieved.

[0079] In one implementation, the user's choice of whether to use an elevator or stairs can be determined by the relationship between the rate of change of air pressure over a preset time period and a preset air pressure difference threshold. For example, if the rate of change of air pressure is greater than the preset air pressure difference threshold, the user is determined to use an elevator; if the rate of change of air pressure is less than the preset air pressure difference threshold, the user is determined to use stairs.

[0080] Because the accelerometer, gyroscope, and barometer operate at low power, they can remain on to record the terminal's acceleration, attitude, and ambient air pressure in real time. The accelerometer, gyroscope, and barometer all acquire information at a 25Hz operating frequency.

[0081] The above technical solution improves the accuracy of determining the retrieval of card information by comprehensively calculating the current usage probability of card information through multi-dimensional feature information.

[0082] In step S12, the feature information is input into the corresponding information calling model to obtain the current usage probability of each identification card information output by the information calling model.

[0083] In this step, there are multiple information retrieval models, and each of the multiple information retrieval models corresponds one-to-one with the multiple identification card information. For example, if the identification card information includes a subway card, a company access card, and a home access card, then the subway card corresponds to one information retrieval model, the company access card corresponds to one information retrieval model, and the home access card corresponds to one information retrieval model.

[0084] refer to Figure 2 As shown, in step S12, the input of the feature information into the corresponding information retrieval model to obtain the current usage probability of each piece of identification card information output by the information retrieval model includes:

[0085] In step S121, the feature information in the multiple dimensions is input into each of the information retrieval models to obtain the current usage probability of the corresponding identification card information output by each of the information retrieval models.

[0086] Following the above embodiments, feature information in each dimension is input into the information retrieval model corresponding to the subway card, resulting in the current usage probability of the subway card output by the information retrieval model. Similarly, feature information in each dimension is input into the information retrieval model corresponding to the company access card, resulting in the current usage probability of the company access card output by the information retrieval model. Likewise, feature information in each dimension is input into the information retrieval model corresponding to the home access card, resulting in the current usage probability of the home access card output by the information retrieval model.

[0087] Based on the above embodiments, each information retrieval model includes a sub-model corresponding to each dimension. During the training process, the sub-model can determine the normal distribution model of the sample feature information corresponding to the sub-model in the model training samples.

[0088] For example, when there are multiple dimensions of feature information, including features in the time dimension and features in the geographical location dimension, the information retrieval model corresponding to the subway card includes sub-models in the time dimension and sub-models in the geographical location dimension. At the same time, the information retrieval model corresponding to the company access card also includes sub-models in the time dimension and sub-models in the geographical location dimension, and the information retrieval model corresponding to the home access card also includes sub-models in the time dimension and sub-models in the geographical location dimension.

[0089] Among them, the sub-model can be a Gaussian mixture model, and in the information retrieval model corresponding to different identification card information, the sample feature information of the sub-model in the same dimension is different during the training process. In the information retrieval model corresponding to the same identification card information, the sample feature information of the sub-model in different dimensions is also different during the training process.

[0090] For example, during the training process, the sample feature information of each sub-model in the identification card information is the feature information of the usage scenario corresponding to that identification card information. For instance, for the time dimension sub-model in the information retrieval model corresponding to the subway card, the sample feature information of this sub-model during the training process is the timestamp of the subway ride scenario corresponding to the subway card information. For the geographical location dimension sub-model in the information retrieval model corresponding to the subway card, the sample feature information of this sub-model during the training process is the geographical location information of the subway ride scenario corresponding to the subway card information.

[0091] Similarly, for the time-dimension sub-model in the information retrieval model corresponding to the company access card, the sample feature information of this sub-model during the training process is the timestamp of the company entry scenario corresponding to the company access card information. For the geolocation-dimension sub-model in the information retrieval model corresponding to the company access card, the sample feature information of this sub-model during the training process is the location information of the company corresponding to the company access card.

[0092] refer to Figure 3 As shown, in step S121, the step of inputting the feature information of the multiple dimensions into each of the information retrieval models to obtain the current usage probability of the corresponding identification card information output by each of the information retrieval models includes:

[0093] In step S1211, for any of the information retrieval models, the feature information of each dimension is input into the corresponding sub-model in the information retrieval model to obtain the usage sub-probability output by each sub-model.

[0094] The usage probability P corresponding to each sub-model can be calculated using the following formula. k :

[0095]

[0096] Among them, f k For any dimension k, σ k μ represents the variance of the sample feature information in this dimension. k This represents the average value of the sample feature information in this dimension.

[0097] In one implementation, since the sub-model trained from the model training samples may not satisfy the standard normal distribution, it is necessary to adjust the trained model in order to obtain a standard normal distribution model of the sample feature information of the sub-model.

[0098] For example, the trained model can be tuned by adjusting parameters:

[0099]

[0100] Among them, R k These are the adjustment parameters used to adjust the sub-models of the training model to a standard normal distribution model based on the information obtained from the training.

[0101] In step S1212, the current usage probability of the identification card information corresponding to the information call model is determined based on the usage sub-probability output by each sub-model.

[0102] For example, the product of the usage probabilities of each sub-model in the information retrieval model is calculated to obtain the current usage probability P output by the information retrieval model corresponding to the identification card information. i,k :

[0103] P i,k =P1*P2*P3…P k

[0104] In one implementation, the weight of each type of information calling the model is determined by pre-setting the observation probability of each identification card information:

[0105] P i,k =a i (P1*P2*P3…P k )

[0106] Among them, a i The preset weights for the information retrieval model are used to characterize the observation probability parameters of the information retrieval model when calculating the current usage probability.

[0107] In step S13, the first target identification card information with the highest current usage probability is retrieved from multiple identification card information.

[0108] Among them, the one with the highest current usage probability indicates that the identification card information has the highest usage probability, so the identification card information is used as the first target identification card information to be called.

[0109] The above technical solution obtains the current usage probability of each identification card information output by the information calling model by inputting the feature information of the acquired electronic device in multiple dimensions into the corresponding information calling model; and calls the first target identification card information with the highest current usage probability among multiple identification card information. This can avoid relying on setting the calling time period and calling location range for identification card information calling, and also improves the accuracy of calling identification card information.

[0110] Based on the above embodiments, referring to Figure 4 As shown, the method further includes:

[0111] In step S41, when the electronic device is successfully identified by the identifier, sample feature information of the electronic device in multiple dimensions and sample identification card information of the successfully identified device are obtained.

[0112] For example, when an electronic device successfully recognizes any identification card information, it acquires the sample feature information of the electronic device in multiple dimensions corresponding to that identification card information, as well as the identification card information of the sample that was successfully recognized by the current identifier.

[0113] Optionally, the sample feature information can be artificially constructed feature information or feature information uploaded by the user in the past.

[0114] In step S42, the sample identification card information is used as the data label of the sample feature information to obtain the model training sample.

[0115] In one possible implementation, the number of model training samples is kept at a certain number. After obtaining model training samples by using sample identification card information as data labels for sample feature information, the number of model training samples is determined. If the number exceeds a preset threshold, the constructed sample feature information that is far removed in time or has low similarity to the current sample identification card information is removed.

[0116] In step S43, the model is invoked based on the training information from the model training samples.

[0117] By adopting the above technical solution, the model training samples can be updated based on the feature information of multiple dimensions that the recognizer has successfully identified, thereby continuously improving the accuracy of information retrieval from the model.

[0118] exist Figure 1 Based on, refer to Figure 5 As shown, the method further includes:

[0119] In step S14, if the verification of the first target identification card information fails, the user is prompted to manually switch the identification card information.

[0120] For example, the user can be prompted to manually switch the identification card information via voice messages, display messages, or a combination of both.

[0121] In step S15, the manually switched identification card information is used as the second target identification card information.

[0122] Based on the above embodiments, the method further includes:

[0123] Based on the feature information corresponding to the verification failure, the sample feature information in the information retrieval model corresponding to the first target identification card information is removed to update the model training samples in the information retrieval model corresponding to the first target identification card information, and the sample feature information in the information retrieval model corresponding to the second target identification card information is added to update the model training samples in the information retrieval model corresponding to the second target identification card information.

[0124] In one possible implementation, based on the feature information corresponding to the verification failure, one or more sample feature information with the highest similarity to the feature information are found, and one or more sample feature information with the highest similarity to the feature information are removed from the sample feature information in the information retrieval model corresponding to the first target identification card information to obtain new model training samples. The average value and variance of the sample feature information are calculated based on the new model training samples, and then the usage probability corresponding to the sub-model is calculated in subsequent information retrieval using the new average value and variance.

[0125] For example, from the sub-model of the information retrieval model corresponding to the first target identification card information, one or more sample feature information with the highest similarity to the feature information is removed to obtain the new model training sample corresponding to the sub-model. Based on the new model training sample, the new mean and variance of the model training sample in the sub-model are calculated to complete the update of the model training sample of the sub-model of the information retrieval model corresponding to the first target identification card information.

[0126] Similarly, based on the feature information corresponding to the verification failure, one or more sample feature information is constructed, and the constructed sample feature information is added to the corresponding information retrieval model to obtain new model training samples. The mean and variance of the sample feature information are calculated based on the new model training samples, and then the sub-probability of the sub-model is calculated in subsequent information retrievals using the new mean and variance.

[0127] For example, one or more sample feature information is added to the sub-model of the information retrieval model corresponding to the user-selected identification card information to obtain the new model training sample corresponding to the sub-model. Based on the new model training sample, the new mean and variance of the model training sample in the sub-model are calculated to complete the update of the model training sample of the sub-model of the information retrieval model corresponding to the second target identification card information.

[0128] The above technical solution allows for the use of training samples to train the information retrieval model corresponding to verification failures and the information retrieval model corresponding to manually selected identification cards during the verification and recognition process. This enables the information retrieval model to learn and improves the accuracy of information retrieval.

[0129] It can be noted that if the same identification card information fails to be verified a large number of times, it indicates that there is a significant error in the information retrieval model corresponding to that identification card information. Therefore, based on the above embodiments, the method further includes:

[0130] If the number of verification failures for the same identification card information exceeds a preset threshold, the information retrieval model corresponding to the identification card information is retrained based on the updated model training samples, and the adjustment parameters and observation parameters of the information retrieval model are updated.

[0131] The adjustment parameters are used to adjust the sub-model of the retrained information retrieval model to a standard normal distribution model, and the observation parameters are the weights of the retrained information retrieval model when calculating the current usage probability.

[0132] For example, with a preset threshold of 100 times, the mean and variance of the model training samples after the 100th time of removing sample feature information are calculated to obtain the mean and variance of the new sample feature information. The model is then retrained based on the sample feature information to obtain the normal distribution model of the sub-model. Based on the new normal distribution model, new adjustment parameters for adjusting the sub-model are determined. At the same time, the observation parameters of the model that are used to adjust the information are determined.

[0133] Optionally, if the number of verification failures for the same identification card information exceeds a preset threshold, the user is prompted to upload feature information within the preset threshold to the cloud. The cloud server then retrains the information retrieval model corresponding to the identification card information and updates the retrained model on other terminals through transfer learning. For example, the information retrieval models on other terminals of the user are updated simultaneously, or the information retrieval models on other users' terminals are updated.

[0134] Based on the same inventive concept, this disclosure also provides an information retrieval device 100 for executing the steps of the information retrieval method provided in the above method embodiments. The device 100 can implement the information retrieval method in software, hardware, or a combination of both. Figure 6 This is a block diagram illustrating an information retrieval device 100 according to an exemplary embodiment. Referring to Figure 6, the device 100 includes: an acquisition module 110, an input module 120, and a retrieval module 130.

[0135] The acquisition module 110 is configured to acquire feature information of the electronic device in multiple dimensions in response to the identification card information call operation, wherein the electronic device stores multiple identification card information.

[0136] The input module 120 is configured to input the feature information into the corresponding information calling model to obtain the current usage probability of each piece of identification card information output by the information calling model;

[0137] The calling module 130 is configured to call the first target identification card information with the highest current usage probability among the plurality of identification card information.

[0138] The aforementioned device obtains the current usage probability of each identification card information output by the information calling model by inputting the acquired feature information of the electronic device in multiple dimensions into the corresponding information calling model; it calls the first target identification card information with the highest current usage probability among multiple identification card information, which can avoid relying on setting the calling time period and calling location range for identification card information calling, and also improves the accuracy of calling identification card information.

[0139] Optionally, the input module 120 is further configured to:

[0140] When the electronic device is successfully identified by the identifier, the sample feature information of the electronic device in the multiple dimensions and the sample identification card information of the successfully identified device are obtained.

[0141] The sample identification card information is used as the data label for the sample feature information to obtain the model training samples;

[0142] The information retrieval model is trained based on the training samples of the model.

[0143] Optionally, there are multiple information retrieval models, and each of the multiple information retrieval models corresponds one-to-one with the multiple identification card information. The input module 120 is configured to input the feature information in the multiple dimensions into each of the information retrieval models respectively, so as to obtain the current usage probability of the corresponding identification card information output by each of the information retrieval models.

[0144] Optionally, each of the information retrieval models includes a sub-model corresponding to each of the dimensions, and the sub-model can determine the normal distribution model of the sample feature information corresponding to the sub-model in the model training samples during the training process;

[0145] The input module 120 is configured to input feature information in each dimension into the corresponding sub-model of the information retrieval model for any of the information retrieval models, and obtain the usage sub-probability output by each sub-model.

[0146] Based on the usage sub-probability output by each sub-model, the current usage probability of the identification card information corresponding to the information call model is determined.

[0147] Optionally, the feature information in the multiple dimensions includes at least two of the following: the location information of the electronic device when responding to the identification card information call operation; the timestamp when responding to the identification card information call operation; the posture information of the electronic device when responding to the identification card information call operation; the motion state information of the user carrying the electronic device within a preset time period before responding to the identification card information call operation; and the up-and-down status information of the user carrying the electronic device within the preset time period before responding to the identification card information call operation.

[0148] Optionally, the device 100 further includes a prompting module configured to prompt the user to manually switch the identification card information if the verification of the first target identification card information fails;

[0149] The calling module 130 is also configured to call the manually switched identification card information as the second target identification card information.

[0150] Optionally, the device 100 further includes an update module, configured to remove sample feature information in the information retrieval model corresponding to the first target identification card information based on the feature information corresponding to the verification failure, so as to update the model training samples in the information retrieval model corresponding to the first target identification card information, and to add sample feature information in the information retrieval model corresponding to the second target identification card information, so as to update the model training samples in the information retrieval model corresponding to the second target identification card information.

[0151] Optionally, the update module is further configured to, when the number of verification failures for the same identification card information exceeds a preset threshold, retrain the information retrieval model corresponding to the identification card information based on the updated model training samples, and update the adjustment parameters and observation parameters of the information retrieval model;

[0152] The adjustment parameters are used to adjust the sub-model of the retrained information retrieval model to a standard normal distribution model, and the observation parameters are the weights of the retrained information retrieval model when calculating the current usage probability.

[0153] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0154] Furthermore, it is worth noting that, for the sake of convenience and brevity, the embodiments described in the specification are all preferred embodiments, and the parts involved are not necessarily essential to the present invention. For example, the input module 120 and the calling module 130 can be independent devices or the same device in specific implementations, and this disclosure does not limit them.

[0155] An electronic device is also provided according to embodiments of this disclosure, including:

[0156] processor;

[0157] Memory used to store processor-executable instructions;

[0158] The processor is configured as follows:

[0159] In response to an identification card information retrieval operation, feature information of the electronic device in multiple dimensions is obtained, wherein the electronic device stores multiple identification card information;

[0160] The feature information is input into the corresponding information retrieval model to obtain the current usage probability of each piece of identification card information output by the information retrieval model.

[0161] The first target identification card information with the highest current usage probability is retrieved from the multiple identification card information.

[0162] According to embodiments of this disclosure, a computer-readable storage medium is also provided, having stored thereon computer program instructions that, when executed by a processor, implement the steps of any of the methods described above.

[0163] Figure 7This is a block diagram illustrating an apparatus 800 for information retrieval according to an exemplary embodiment. For example, apparatus 800 may be a mobile phone, computer, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0164] Reference Figure 7 The device 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input / output (I / O) interface 812, a sensor component 814, and a communication component 816.

[0165] Processing component 802 typically controls the overall operation of device 800, such as operations associated with display and identification card information entry. Processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the aforementioned information retrieval method. Furthermore, processing component 802 may include one or more modules to facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.

[0166] Memory 804 is configured to store various types of data to support the operation of device 800. Examples of this data include instructions for any application or method operating on device 800, acquiring characteristic information such as acceleration information, responding to call operations, etc. Memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0167] The power supply component 806 provides power to the various components of the device 800. The power supply component 806 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the device 800.

[0168] Multimedia component 808 includes a screen that provides an output interface between the device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 808 includes a front-facing camera and / or a rear-facing camera. When the device 800 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0169] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals when device 800 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 804 or transmitted via communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

[0170] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0171] Sensor assembly 814 includes one or more sensors for providing status assessments of various aspects of device 800. For example, sensor assembly 814 may detect the on / off state of device 800, the relative positioning of components such as the display and keypad of device 800, changes in the position of device 800 or a component of device 800, the presence or absence of user contact with device 800, the orientation or acceleration / deceleration of device 800, and temperature changes of device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 814 may also include an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor.

[0172] Communication component 816 is configured to facilitate wired or wireless communication between device 800 and other devices. Device 800 can access wireless networks based on communication standards, such as WiFi, 2G, 3G, 4G, or 5G, or combinations thereof. In one exemplary embodiment, communication component 816 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 816 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0173] In an exemplary embodiment, the device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the information retrieval method described above.

[0174] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, which can be executed by a processor 820 of the device 800 to complete the aforementioned information retrieval method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0175] In another exemplary embodiment, a computer program product is also provided, the computer program product comprising a computer program executable by a programmable device, the computer program having a code portion for performing the above-described information invocation method when executed by the programmable device.

[0176] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of this disclosure. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0177] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. An information calling method characterized by comprising: The method comprises: In response to an identification card information calling operation, obtaining feature information of the electronic device in multiple dimensions, wherein the electronic device stores multiple identification card information; Inputting the feature information into a corresponding information calling model to obtain a current use probability of each identification card information output by the information calling model, wherein the information calling model is multiple, and the multiple information calling models correspond to the multiple identification card information one by one, each information calling model comprises a sub-model corresponding to each dimension, the sample feature information of the sub-model corresponding to the same dimension in the information calling model corresponding to different identification card information is different in the training process, and the sample feature information of the sub-model corresponding to different dimensions in the information calling model corresponding to the same identification card information is different in the training process; Calling a first target identification card information with the largest current use probability in the multiple identification card information.

2. The method of claim 1, wherein, The method further comprises: When the electronic device is successfully identified by an identifier, obtaining sample feature information of the electronic device in the multiple dimensions and sample identification card information successfully identified by the identifier; Taking the sample identification card information as a data label of the sample feature information to obtain a model training sample; Training the information calling model according to the model training sample.

3. The method of claim 1, wherein, The method further comprises: Inputting the feature information in the multiple dimensions into each information calling model to obtain the current use probability of the corresponding identification card information output by each information calling model.

4. The method of claim 3, wherein, The sub-model can determine a normal distribution model of the sample feature information corresponding to the sub-model in the model training sample in the training process. The method further comprises: For any information calling model, inputting the feature information in each dimension into the corresponding sub-model in the information calling model to obtain a use sub-probability output by each sub-model; Determining the current use probability of the identification card information corresponding to the information calling model according to the use sub-probability output by each sub-model.

5. The method of claim 1, wherein, The feature information in the multiple dimensions comprises at least two of the following: position information of the electronic device in response to an identification card information calling operation, a timestamp in response to an identification card information calling operation, posture information of the electronic device in response to an identification card information calling operation, motion state information of a user carrying the electronic device within a preset time period before the identification card information calling operation, and up-and-down state information of the user carrying the electronic device within the preset time period before the identification card information calling operation.

6. The method according to any one of claims 1-5, characterized in that, The method further comprises: In the case that the first target identification card information fails to be verified, prompting a user to manually switch the identification card information; Taking the identification card information switched manually as a second target identification card information for calling.

7. The method of claim 6, wherein, The method further comprises: According to the feature information corresponding to the verification failure, the sample feature information in the information calling model corresponding to the first target identification card information is eliminated to update the model training sample in the information calling model corresponding to the first target identification card information, and the sample feature information in the information calling model corresponding to the second target identification card information is added to update the model training sample in the information calling model corresponding to the second target identification card information.

8. The method of claim 7, wherein, The method further comprises: In the case that the number of verification failures of the same identification card information exceeds a preset number threshold, the information calling model corresponding to the identification card information is retrained according to the updated model training sample, and the adjustment parameter and the observation parameter of the information calling model are updated; The adjustment parameter is used to adjust the sub-model of the retrained information calling model to a standard normal distribution model, and the observation parameter is the weight of the retrained information calling model in calculating the current use probability.

9. An information calling apparatus characterized by comprising: Comprise: An acquisition module configured to acquire feature information of an electronic device in multiple dimensions in response to an identification card information calling operation, wherein the electronic device stores multiple identification card information; An input module configured to input the feature information into a corresponding information calling model to obtain a current use probability of each identification card information output by the information calling model, wherein the information calling model is multiple, and the multiple information calling models correspond to the multiple identification card information one by one, each information calling model includes a sub-model corresponding to each dimension, the sample feature information of the sub-model corresponding to the same dimension in the information calling model corresponding to different identification card information is different, and the sample feature information of the sub-model corresponding to different dimensions in the information calling model corresponding to the same identification card information is different; A calling module configured to call a first target identification card information with the largest current use probability in the multiple identification card information.

10. An electronic device, comprising: Comprise: A processor; A memory for storing processor-executable instructions; The processor is configured to: Acquire feature information of an electronic device in multiple dimensions in response to an identification card information calling operation, wherein the electronic device stores multiple identification card information; Input the feature information into a corresponding information calling model to obtain a current use probability of each identification card information output by the information calling model, wherein the information calling model is multiple, and the multiple information calling models correspond to the multiple identification card information one by one, each information calling model includes a sub-model corresponding to each dimension, the sample feature information of the sub-model corresponding to the same dimension in the information calling model corresponding to different identification card information is different, and the sample feature information of the sub-model corresponding to different dimensions in the information calling model corresponding to the same identification card information is different; Call a first target identification card information with the largest current use probability in the multiple identification card information.

11. A computer-readable storage medium having stored thereon computer program instructions, wherein, The program instructions, when executed by the processor, implement the steps of the method of any one of claims 1-8.

Images