Information processing method and system, electronic equipment and storage medium

Abstract

The invention provides an information processing method and system, electronic equipment and a storage medium, and the method comprises the steps: determining a target region based on a fingertip movement track of a user on a page when a circle matching function is started; identifying contents in the target area based on an OCR technology and / or an image identification technology, and analyzing by using context data of the user to determine target data of the user; analyzing the target data by adopting a prediction model, and predicting demands or problems of the user; replying according to requirements or questions of the user and displaying a reply result to the user; in other words, the circling function and the matching function are linked; determining an area where the user needs to answer based on the fingertip movement track of the user, and replying to a predicted demand or question corresponding to the area; problems caused by the mode that a user looks up information or solves the problems depending on manual input and page navigation are avoided, the information processing efficiency is improved, the method is visual and convenient, and the requirement of the user for an efficient interaction mode is met.

Description

Technical Field

[0001] The present invention belongs to the field of data processing technology, and more specifically, particularly relates to an information processing method and system, electronic equipment, and storage medium. Background Art

[0002] With the continuous advancement and popularization of mobile Internet technology, mobile banking has become an indispensable part of people's daily lives. According to statistics, the number of mobile banking users has exceeded 450 million, and the number of monthly active users has reached 200 million. As the user scale continues to expand, user behaviors and attributes are becoming more and more diverse. Correspondingly, users' requirements for mobile banking matching functions are also increasing, especially when new products are released, such as popular activities such as commemorative coin reservations, the pressure on matching functions is even more significant. For example, during a new product release, the matching function TPM (transactions per second) once reached 120,000 times, fully demonstrating users' urgent need for efficient matching functions.

[0003] However, in current mobile banking applications, users mainly rely on manual input and page navigation to find information or solve problems.

[0004] However, this method is not only inefficient, but also lacks intuitiveness and convenience, and it is difficult to meet users' needs for efficient interaction methods. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide an information processing method and system, an electronic device, and a storage medium, which are used to improve the efficiency of information processing, are intuitive and convenient, and meet the user's demand for efficient interaction methods.

[0006] The first aspect of the present application discloses an information processing method, comprising:

[0007] When the circle matching function is turned on, the target area is determined based on the user's fingertip movement trajectory on the page;

[0008] Identify the content in the target area based on OCR technology and / or image recognition technology, and analyze using the user's context data to determine the user's target data;

[0009] Analyzing the target data using a prediction model to predict the needs or problems of the user;

[0010] According to the needs or questions of the user, a reply is made and the reply result is displayed to the user.

[0011] Optionally, the OCR technology and / or image recognition technology is used to identify the content in the target area, including:

[0012] If the content in the target area includes text, using the OCR technology to recognize the text in the target area;

[0013] If the content in the target area includes an image or an interface element, the image recognition technology is used to recognize the image or the interface element in the target area.

[0014] Optionally, replying according to the needs or questions of the user and displaying the reply result to the user includes:

[0015] Searching for the needs or questions of the user to obtain search results corresponding to the needs or questions;

[0016] The search results corresponding to the requirements or questions are displayed to the user.

[0017] Optionally, replying according to the needs or questions of the user and displaying the reply result to the user includes:

[0018] Transmitting the user's needs or questions to customer service, and having customer service respond to the user's needs or questions;

[0019] The customer service responds to the request or question and displays it to the user.

[0020] Optionally, the context data includes: user behavior history, transaction path, location information and situation awareness data; wherein the situation awareness data is data perceived in combination with the user's current time, date and environmental factors.

[0021] Optionally, the process of building and updating the prediction model includes:

[0022] Collect user behavior data;

[0023] Standardizing the time series data in the user's behavior data to obtain training samples;

[0024] The training samples are trained using an LSTM model to learn the time series pattern of user behavior and obtain a prediction model of user behavior;

[0025] The prediction model is updated regularly based on the prediction results and user behavior data.

[0026] Optionally, when the circle matching function is turned on, before determining the target area based on the fingertip movement trajectory of the user on the page, the method further includes:

[0027] When it is detected that the number and amplitude of shaking of the mobile phone within the preset time meet the preset conditions, it is determined that the circle matching function is turned on.

[0028] The second aspect of the present application discloses an information processing system, comprising:

[0029] The circle unit is used to determine the target area based on the user's fingertip movement trajectory on the page when the circle matching function is turned on;

[0030] An identification unit, configured to identify the content in the target area based on OCR technology and / or image recognition technology, and analyze the user's context data to determine the user's target data;

[0031] A prediction unit, configured to analyze the target data using a prediction model to predict the needs or problems of the user;

[0032] The reply unit is used to reply according to the needs or questions of the user and display the reply result to the user.

[0033] The third aspect of the present application discloses an electronic device, comprising:

[0034] one or more processors;

[0035] a storage device having one or more programs stored thereon;

[0036] When one or more programs are executed by one or more processors, the one or more processors implement the information processing method as described in any one of the first aspects of the present application.

[0037] A fourth aspect of the present application discloses a storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the information processing method as described in any one of the first aspects of the present application is implemented.

[0038] It can be seen from the above technical scheme that an information processing method provided by the present invention includes: when the circle matching function is turned on, based on the user's fingertip movement trajectory on the page, determining the target area; based on OCR technology and / or image recognition technology, identifying the content in the target area, and using the user's context data for analysis to determine the user's target data; using a predictive model to analyze the target data and predict the user's needs or questions; based on the user's needs or questions, replying and displaying the reply results to the user; that is, linking the circle function with the matching function; determining the area that the user needs to answer based on the user's fingertip movement trajectory, and replying to the predicted needs or questions corresponding to the area; avoiding problems caused by users relying on manual input and page navigation to find information or solve problems, improving the efficiency of information processing, being intuitive and convenient, and meeting users' needs for efficient interaction methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 is a flow chart of an information processing method provided by an embodiment of the present invention;

[0041] Figure 2 is a flow chart of another information processing method provided by an embodiment of the present invention;

[0042] Figure 3 is a schematic diagram of an information processing system provided by an embodiment of the present invention;

[0043] Figure 4 is a schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] In the present application, the term "comprise", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of more restrictions, the elements defined by the sentence "comprise one ..." do not exclude the existence of other identical elements in the process, method, article or equipment including the elements. In addition, the terms "first", "second", "third", "fourth", etc. (if present) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable in appropriate circumstances, so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here, for example.

[0046] The embodiment of the present application discloses an information processing method for solving the problem in the prior art that users mainly rely on manual input and page navigation to find information or solve problems. This method is not only inefficient, but also lacks intuitiveness and convenience, and it is difficult to meet the user's demand for efficient interaction methods.

[0047] Shake-to-Activate is a smartphone interaction method that activates specific functions or interfaces within an application by detecting the shaking action of the device.

[0048] Circle Drawing: On a touch screen device, the user draws a circle with his finger or stylus to highlight or select a specific area or content on the screen.

[0049] OCR (Optical Character Recognition), a technology that converts printed or handwritten text into machine-encoded text, is used in the present invention to recognize and extract text information from the content circled by the user.

[0050] Image Recognition, a branch of computer vision, is used to identify and process objects, scenes, text and other information in images.

[0051] Contextual Data refers to information such as the user's behavioral background, historical transaction records, current location, etc., which is used to help understand the user's specific situation and needs.

[0052] LSTM (Long Short-Term Memory), a special type of recurrent neural network (RNN), is very effective in processing time series data and learning long-term dependencies.

[0053] Machine Learning, a branch of AI, refers to the ability of computer systems to use algorithms and statistical models to learn from data and make decisions or predictions.

[0054] Multimodal data refers to data that combines multiple types and formats, such as text, images, sounds, etc., to provide richer information to support decision-making and analysis.

[0055] Intelligent Prediction is the process of using algorithms and models to predict future events or user behaviors based on existing data.

[0056] See also Figure 1 , the information processing method comprises:

[0057] S101. When the circle matching function is turned on, a target area is determined based on the movement trajectory of the user's fingertips on the page.

[0058] Specifically, first, it is necessary to detect whether the user has turned on the circle search function. This is usually achieved through user interface settings or button clicks.

[0059] Second, once the feature is turned on, start tracking the user's fingertips on the page. This is usually achieved through touch screen technology, which can capture the movement path of the user's fingertips on the screen.

[0060] Based on the fingertip motion trajectory, the system can identify the area circled by the user's finger. This area is usually regarded as the target area that the user is interested in or wants to query.

[0061] Of course, the fingertip motion trajectory can be regarded as a trajectory that constitutes a closed area (the trajectory ends are connected), of course, a semi-open trajectory (the first part of the trajectory is close and wrapped), or a fully open line area (the first part of the trajectory is far away).

[0062] S102: Identify the content in the target area based on OCR technology and / or image recognition technology, and analyze the user's context data to determine the user's target data.

[0063] Specifically, for text content, the system uses OCR (optical character recognition) technology to convert text in images into editable and searchable text formats. For non-text content (such as pictures, icons, etc.), the system uses image recognition technology to identify and analyze image elements in the target area. The system collects user contextual data, such as historical search records, browsing habits, geographic location, etc. This data is used to further analyze the content in the target area to determine the specific information that the user may want to query or know. Combining OCR / image recognition results and contextual data analysis, the system determines the user's target data. This data may be text, pictures, or other forms of information.

[0064] S103: Analyze target data using a prediction model to predict user needs or problems.

[0065] Select a suitable prediction model, which can be pre-built.

[0066] The target data is passed as input to the prediction model; of course, the model may also need to input other data, such as the user's contextual data, historical behavior data, etc. The prediction model analyzes the input data and outputs a prediction of the user's needs, which may be a specific query, question, or intent.

[0067] S104: Reply according to the user's needs or questions and display the response results to the user.

[0068] Based on the predicted user needs or questions, the corresponding response is automatically generated. This response may be a simple answer, a detailed explanation, a link to a page with relevant information, etc.

[0069] The generation of a response may involve multiple steps, such as information retrieval, text generation, etc.

[0070] The system displays the generated response results to the user. This is usually achieved through a user interface, such as popping up a dialog box, displaying a prompt box on the page, etc.

[0071] Responses should be presented in a way that is clear, understandable, and consistent with user expectations.

[0072] Users may give feedback on the reply results, such as liking, commenting, or asking new questions. This feedback can be collected and used to improve the prediction model and reply generation algorithm.

[0073] Through the above steps, the information processing method can predict the user's needs or questions based on the user's fingertip motion trajectory and context data, and give a corresponding response. This method has broad application prospects in the fields of smart devices, mobile applications, etc.

[0074] In this embodiment, when the circle matching function is turned on, the target area is determined based on the user's fingertip movement trajectory on the page; the content in the target area is identified based on OCR technology and / or image recognition technology, and the user's context data is used for analysis to determine the user's target data; a predictive model is used to analyze the target data to predict the user's needs or questions; based on the user's needs or questions, a reply is made and the reply result is displayed to the user; that is, the circle function and the matching function are linked; the area that the user needs to answer is determined based on the user's fingertip movement trajectory, and the predicted needs or questions corresponding to the area are replied; problems caused by users relying on manual input and page navigation to find information or solve problems are avoided, the efficiency of information processing is improved, and it is intuitive and convenient to meet users' needs for efficient interaction.

[0075] Optionally, OCR technology and / or image recognition technology are used to identify the content in the target area, including:

[0076] If the content in the target area includes text, OCR technology is used to recognize the text in the target area; if the content in the target area includes images or interface elements, image recognition technology is used to recognize the images or interface elements in the target area.

[0077] When processing the content in the target area, OCR technology and / or image recognition technology are used to accurately identify its content. The specific recognition process is as follows:

[0078] When the target area contains text content, OCR (Optical Character Recognition) technology is preferred. OCR technology converts the text information in the image into an editable and searchable text format by scanning the target area. This technology can efficiently and accurately recognize text in multiple languages ​​and fonts, facilitating subsequent text processing and analysis.

[0079] If the target area contains images or interface elements (such as buttons, icons, background images, etc.), image recognition technology is used. Image recognition technology extracts image features of the target area and compares and matches them with the pre-trained image database. This technology can identify the image type, objects, and interface elements in the target area, providing support for subsequent image processing, classification, and retrieval.

[0080] It should be noted that image recognition technology is mainly used to analyze the area circled by the user in the mobile banking application interface. This covers the recognition of static image components (such as buttons, icons, and text labels) and dynamic display content (such as transaction details refreshed in real time). Specifically, a convolutional neural network (CNN) can be used to extract image features. With its powerful capabilities, CNN can effectively identify and analyze patterns and structures within images, and then accurately identify the interface elements circled by the user. Subsequently, the system will match the identified image elements with the interface element information pre-stored in the background database to determine the specific functions or information content that the user is concerned about.

[0081] The specific process of data matching is as follows: First, using the preset algorithm, the system compares the identified image feature information with the relevant data in the background service one by one. Secondly, combined with the detailed parameters such as the specific position, size and shape of the image features, the system can accurately match the corresponding interface elements or functional modules.

[0082] In practical applications, OCR technology and image recognition technology can complement each other and jointly improve the accuracy and efficiency of identifying content in the target area.

[0083] For example, when identifying a complex target area containing text and images, you can first use OCR technology to identify the text content, and then use image recognition technology to identify images and interface elements.

[0084] By optimizing the recognition process and algorithm, the accuracy and speed of identifying the content in the target area can be further improved, providing users with a better user experience.

[0085] In summary, OCR technology and image recognition technology play an important role in identifying the content of the target area. They can accurately identify text, images and interface elements, and provide strong support for subsequent text processing, image analysis and interface interaction.

[0086] Optional, contextual data, including: user behavior history, transaction paths, location information, and situational awareness data.

[0087] Among them, situation awareness data is data perceived in combination with the user's current time, date and environmental factors.

[0088] User behavior history: user operation records in the application, such as clicks, slides, page access history, etc.

[0089] Transaction path: records of various transaction activities conducted by users, such as transfers, investments, loans, etc.

[0090] Location information: The user's geographic location data when using the application helps to understand the user's physical environment and possible demand scenarios.

[0091] Situational awareness data: Combines the user's current time, date and other environmental factors, such as surrounding noise levels, lighting conditions, etc., to provide richer contextual information.

[0092] That is to say, by combining various data modalities such as user behavior history, transaction path tracking, and location information, the present invention can perform more comprehensive context analysis and achieve in-depth multimodal data analysis. In order to integrate these multimodal data and provide a comprehensive user context perspective, data fusion technology can be used. In addition, with the help of machine learning algorithms, we can deeply analyze user behavior patterns and preferences, providing strong support for personalized services.

[0093] In summary, by using precise image recognition technology and comprehensive multimodal data analysis, this application can gain a deeper insight into and understand the actual needs of users in mobile banking applications, thereby providing users with a more personalized and efficient service experience.

[0094] In this embodiment, OCR technology is combined to recognize text, image recognition technology is used to process circled content, and user context data (such as transaction path, behavior data, location information) is used for comprehensive analysis to achieve multimodal data integration and improve prediction accuracy.

[0095] Optionally, respond based on the user's needs or questions and display the response results to the user, including:

[0096] Search for the user's needs or questions to obtain search results corresponding to the needs or questions; and display the search results corresponding to the needs or questions to the user.

[0097] Specifically, accurate searches are performed based on user needs or questions, and relevant search results are obtained. First, the user's needs or questions need to be accurately parsed. This includes understanding the text or image content entered by the user, and identifying the keywords, context, and intent therein. Secondly, based on the parsed needs or questions, the system performs accurate search operations. This may involve accessing internal databases, calling external APIs, using search engines, and other methods. Next, the content most relevant to the user's needs or questions is filtered out from the search results. This is usually sorted and filtered based on multiple dimensions such as relevance, authority, and timeliness.

[0098] Display search results corresponding to needs or questions to users in a user-friendly way. Design a clear and intuitive display interface for users to present search results. This interface should be easy to navigate and convenient for users to quickly find the content they are interested in. Display filtered and sorted search results to users in the form of lists, cards, summaries, etc. For each search result, the system can provide detailed information such as title, summary, link, etc. for users to learn more. In addition to displaying search results, some interactive functions can also be provided, such as clicking links to view detailed content, liking or commenting on search results, etc. These functions can enhance user participation and satisfaction.

[0099] Through the above steps, the system can perform accurate searches based on the user's needs or questions, and present relevant search results to the user in a user-friendly manner. This not only improves the practicality and convenience of the system, but also enhances the user's experience and satisfaction.

[0100] Optionally, respond based on the user's needs or questions and display the response results to the user, including:

[0101] The user's needs or questions are transmitted to customer service, and customer service responds to the user's needs or questions; the information provided by customer service based on the needs or questions is displayed to the user.

[0102] First, the user's needs or questions will be transmitted to the customer service department in a timely and accurate manner. This step ensures that the user's questions can be quickly received by the customer service staff, laying the foundation for subsequent response work.

[0103] After receiving the user's needs or questions, the customer service staff will respond to the user's questions in detail based on their professional knowledge and experience. This response process is aimed at accurately understanding the user's intentions and providing answers or solutions that meet their needs.

[0104] After completing the reply, the customer service staff will transmit the reply information back to the client through the system. The client will then present this information to the user in a clear and intuitive manner to ensure that the user can easily understand and obtain the answers or solutions they need.

[0105] In summary, this process covers the entire process from receiving user needs to displaying reply information, aiming to provide users with an efficient and accurate customer service experience.

[0106] It should be noted that when faced with a problem, users need to write it down and communicate with customer service. This process usually involves tedious operations and waiting, which reduces the efficiency of interaction.

[0107] In this embodiment, after turning on the circle matching function, the user can directly circle the problem on the page where the problem occurs. After completing the circle, the user is directly connected to the customer service, which avoids the user from checking back and forth, provides answers or help in a timely manner, and improves the interaction efficiency.

[0108] Optionally, the process of building and updating a predictive model includes:

[0109] Collect user behavior data; standardize the time series data in the user behavior data to obtain training samples; use the LSTM model to train the training samples, learn the time series pattern of user behavior, and obtain a prediction model for user behavior; regularly update the prediction model based on the prediction results and user behavior data.

[0110] That is to say, Figure 2 As shown in Figure 1, the data collection phase. The first step is to comprehensively collect user behavior data. This data may cover a variety of user behaviors such as clicks, browsing, purchases, comments, and the timestamps of these behaviors. The data source can be website logs, application logs, user feedback, etc., to ensure the authenticity and integrity of the data.

[0111] Data preprocessing stage. Time series data standardization: Since user behavior data is usually presented in the form of time series, these data need to be standardized. Standardization may include operations such as missing value filling, outlier processing, and data smoothing to ensure data consistency and stability. In addition, time series data needs to be converted into a format suitable for model training, such as dividing the time series into training sets, validation sets, and test sets.

[0112] That is to say, we collect all kinds of user behavior data in mobile banking applications, including click operations, sliding actions, and timestamps when visiting different pages. Then, we preprocess these data and convert them into a format suitable for long short-term memory network (LSTM) processing. Specifically, we organize these behavior data into serialized time series data.

[0113] Before time series data is used for model training, it is standardized and normalized to ensure that the input data falls within the appropriate range that the model can process efficiently. The process of processing time series data for standardization and normalization usually involves standardization, normalization, and processing of time series data.

[0114] The purpose of standardization is to convert the data into a distribution with a mean of 0 and a standard deviation of 1, thereby eliminating scale differences in the data and making different features comparable. This is very important for many machine learning algorithms, especially distance-based algorithms (such as K-nearest neighbors, support vector machines) and gradient descent algorithms. The formula for standardization is as follows:

[0115]

[0116] Where: x is the original data point; μ is the mean of the data; σ is the standard deviation of the data; z is the standardized data point.

[0117] Normalization is scaling the data to a specific range (usually 0 to 1 or -1 to 1). This process is very important for algorithms that are sensitive to the absolute size of the data (such as neural networks), which can speed up the training process and improve model performance. A common method of normalization is minimum-maximum scaling, with the following formula:

[0118]

[0119] Where: x is the original data point; x min is the minimum value in the data; x max is the maximum value in the data; x norm are the normalized data points.

[0120] In the time series data processing flow, the following steps are usually performed in sequence:

[0121] First, data cleaning is performed. This step mainly solves the problems of missing values ​​and outliers in the data to ensure the integrity and accuracy of the data. Then, the features and labels are separated. In the scenario of supervised learning, time series data will be split into features (usually past observation records) and labels (usually future values ​​to be predicted). Next, the feature data is standardized or normalized. The specific method is selected according to the characteristics of the data, aiming to adjust the data to a range suitable for model processing. At the same time, according to actual needs, the label data may also need to undergo a standardization or normalization process, especially when dealing with regression problems.

[0122] In actual operation, it is worth noting that the parameters used for standardization and normalization (such as mean, minimum value, maximum value, etc.) should be calculated based on the training set data, and these parameters need to be consistently applied to the entire data set (training set and test set) to ensure the consistency of the data during the training and testing stages.

[0123] Model training stage. The LSTM (Long Short-Term Memory) model is selected for training because it has the ability to capture long-term dependencies in time series data. The standardized time series data is used as input, and the time series pattern of user behavior is learned through the LSTM model. During the training process, the model performance is optimized by adjusting the model parameters (such as learning rate, number of hidden layers, etc.) until the predetermined training goal is achieved.

[0124] The collected user behavior data is used to train the LSTM model, aiming to enable the model to capture and learn the time series patterns in user behavior. This process involves multiple training iterations, during which we continuously adjust and optimize the model parameters in order to minimize the prediction error and thus improve the accuracy and performance of the model.

[0125] With the help of the well-trained LSTM model, the system can predict the user’s next possible operation or needs. For example, if the user is accustomed to transferring money after checking the account balance, the model can predict the user’s intention to transfer money after the user browses the balance.

[0126] Model update and optimization stage. Use the test set data to perform prediction evaluation on the trained LSTM model, and evaluate the prediction accuracy of the model by comparing the prediction results with the actual user behavior data. Update the prediction model regularly based on the evaluation results. The update method may include retraining the model, adjusting model parameters, or introducing new feature variables. At the same time, pay attention to the dynamic changes of user behavior data and incorporate new data into model training in a timely manner to maintain the timeliness and accuracy of the model.

[0127] After the model is deployed, it is necessary to continuously monitor the model's predictive performance to promptly detect and address any performance degradation or deviation issues that may occur.

[0128] By collecting user feedback and observing the performance of the model in actual applications, the model structure and parameter settings are continuously optimized to improve the model's predictive ability and user experience.

[0129] That is, as user behavior continues to change, the LSTM model will continue to learn new behavior patterns. The model is updated regularly to ensure that its prediction accuracy improves as user behavior changes.

[0130] In summary, the process of building and updating the prediction model is an iterative process that requires continuous data collection, data preprocessing, model training, and model performance evaluation to ensure that the model can accurately predict user behavior and provide better services to users.

[0131] LSTM, or long short-term memory, is a special type of recurrent neural network (RNN) designed to effectively process and predict data with time series characteristics. Its unique gating mechanism, including forget gate, input gate, and output gate, enables LSTM to control the retention and forgetting of information, thus solving the problem of traditional RNN in dealing with long-term dependencies.

[0132] Forget gate: Responsible for deciding which information should be removed from the cell state. This mechanism looks at the previous hidden state and the current input to decide which information is no longer important, thus ensuring that the network can forget unnecessary information.

[0133] Input gate: Responsible for updating the new information of the cell state. It determines what new information should be added to the cell state to help the network learn new data patterns.

[0134] Output gate: Determines the output value based on the current cell state. This mechanism looks at the cell state and the current input to determine what information should be output, ensuring that the network can produce accurate predictions.

[0135] LSTM model training process:

[0136] When training the LSTM model, the back-propagation algorithm is used to adjust the network weights to minimize the prediction error. This process includes the following key steps:

[0137] Initialize network weights: Randomly initialize the weights of the LSTM network, including the weights and biases of the input gate, forget gate, output gate, and unit state.

[0138] Forward propagation: At each time step, the LSTM network receives input data and processes them. The gating mechanism controls the inflow, retention, and outflow of information, and updates the cell state and hidden state.

[0139] Calculate the error: After the network generates an output, calculate the difference between the output and the actual value (or target value), that is, the error. Usually a loss function (such as mean square error or cross entropy) is used to measure the error size.

[0140] Backpropagation: The error is propagated back to the network using the chain rule. The contribution of each weight to the total error (gradient) is calculated, and the weights are updated using an optimization algorithm such as stochastic gradient descent. The goal of the update is to reduce the total error and improve network performance.

[0141] Iterative optimization: Repeat the process of forward propagation and back propagation until the network performance reaches a predetermined standard or the error is reduced to an acceptable level.

[0142] During the training process, we also need to pay attention to the gradient vanishing and exploding problems. Due to the continuous product operations during backpropagation, the gradient may become very small (vanishing) or very large (exploding). To solve this problem, gating mechanisms and some techniques (such as gradient clipping) can be used to ensure the stability of the model.

[0143] Application of LSTM model in mobile banking APP:

[0144] With the help of the trained LSTM model, the mobile banking app can predict and meet user needs more intelligently and more personally. Specific application scenarios include:

[0145] Integrated search and customer service: Based on the user's circled content and predicted results, the system automatically triggers the corresponding search or connects to customer service. This provides users with a more convenient and efficient way to get help.

[0146] Personalized response: The system provides personalized solutions based not only on the user's current interactive content, but also on the user's overall behavioral background (such as historical transaction records, browsing habits, etc.). This helps improve user satisfaction and loyalty.

[0147] In summary, the present invention provides an efficient, intuitive and intelligent solution for mobile banking users. By using LSTM-based machine learning algorithms to analyze user data, predict possible user needs or problems, and provide targeted search results or customer service support, the user's operating efficiency and experience in the application are significantly improved. As time goes by and data accumulates, the LSTM model will continuously optimize its prediction accuracy and provide users with more accurate and personalized services.

[0148] It should be noted that the interface design of some current mobile banking apps is still static, seriously lacking the ability to dynamically respond to user behavior and needs. This leads to a problem that cannot be ignored: no matter how the user operates or uses the app, its interface always maintains a fixed mode and cannot provide personalized services based on the user's actual situation and needs.

[0149] However, this application completely breaks this limitation by introducing a series of advanced technologies and innovative design concepts. Specifically, it not only innovates in functions, such as the shake-to-activate circle function, the application of OCR and image recognition technology, but also achieves a qualitative leap in interface design.

[0150] The interface design of this application can capture and analyze user behavior data in real time, including user clicks, swipes, dwell time, etc., so as to accurately understand the user's real needs and preferences. Based on this data, our interface can dynamically adjust the layout, color, font and other elements to ensure that users can get the most personalized service that meets their needs in different scenarios.

[0151] For example, when a user frequently accesses a specific function, our interface can intelligently place the function in a more prominent position so that the user can find and use it more quickly. At the same time, we can also recommend related functions or services to the user based on his or her preferences and habits, further improving the user's experience and satisfaction.

[0152] In addition, the interface design also fully considers the characteristics and needs of different user groups. By collecting and analyzing users' basic information, historical behavior and other data, it is possible to tailor an interface style and functional layout that better suits their personality and needs. This personalized design not only improves user satisfaction and loyalty, but also enhances the competitiveness of mobile banking apps.

[0153] In summary, this application completely subverts the static mode of traditional mobile banking APP interface design by introducing the concept of dynamic response and personalized service. Our innovative design not only improves the user experience and satisfaction, but also points out the direction for the future development of mobile banking APP.

[0154] Optionally, when the circle matching function is turned on, before determining the target area based on the user's fingertip movement trajectory on the page, the following is further included:

[0155] When it is detected that the number and amplitude of shaking of the mobile phone within the preset time meet the preset conditions, it is determined that the circle matching function is turned on.

[0156] When the system detects that the number of times the phone is shaken and the amplitude of each shake meet the preset conditions within a preset time period, it will automatically determine that the circle matching function has been activated and turned on.

[0157] That is, the circle matching function can be turned on by shaking. Specifically, when users encounter problems or look for specific functions in the mobile banking app, they can easily activate the circle function by simply shaking the phone. This innovative design allows users to circle or mark the specific area where they encounter problems directly on the screen. The system will then intelligently analyze the area circled by the user and give corresponding replies or suggestions based on this information. This intuitive and efficient method not only improves the user experience, but also makes problem solving faster and more convenient.

[0158] In this embodiment, a shake function is introduced to activate the circle mode, and the user can directly circle the part in question on the screen, simplifying the problem identification process.

[0159] It should be noted that in current mobile banking apps, users mainly rely on traditional methods such as navigation menus, keyword searches, FAQ and help pages, and online customer service to find specific functions or solve problems in transactions. However, these methods have problems such as high time consumption, low efficiency, poor user experience, information overflow, and limited customer service resources.

[0160] In order to overcome these limitations, this embodiment proposes an innovative solution to significantly improve the efficiency of user problem solving and optimize user experience in the following ways:

[0161] Shake to activate circle function: Users can activate circle mode by simply shaking the phone. This intuitive and quick way allows users to easily mark specific problem areas, thereby improving the efficiency of problem identification.

[0162] OCR and image recognition technology: Apply OCR (optical character recognition) and image recognition technology to automatically identify the text and image content circled by the user, reducing the tedious process of users relying on text descriptions, making problem identification more accurate and efficient.

[0163] Contextual data analysis: Comprehensive analysis of multimodal data such as user behavior background and location information is used to provide users with more personalized services. This analysis not only takes into account the user's current needs, but also combines their historical behavior and preferences, thereby improving the pertinence and relevance of the service.

[0164] Intelligent prediction of LSTM model: Through the LSTM (Long Short-Term Memory) model, user behavior and needs are intelligently predicted to provide targeted search results or customer service support. This prediction capability enables the system to gain insight into user needs in advance, reducing the user's search time and effort in the application.

[0165] To sum up, this patent, by combining advanced technology and innovative user interaction methods, not only improves the efficiency of users in finding information and solving problems, but also greatly improves the user experience and reduces dependence on manual customer service resources.

[0166] Compared with the current closest technical solutions, the advantages of the present invention lie in its intuitive, multi-modal interaction and its ability to comprehensively analyze user needs. Although existing technologies such as intelligent chatbots, voice recognition and command execution, personalized recommendation systems, and image recognition applications have improved the user experience and interaction efficiency of banking applications to a certain extent, they are generally unable to achieve comprehensive, accurate, and personalized services like those proposed by the present invention.

[0167] The current technical solutions closest to the present invention are mainly focused on improving the user interaction efficiency and intelligent services of mobile banking applications. These solutions usually also involve the following aspects: 1) Intelligent chatbots: Provide real-time responses by simulating human customer service. These robots are usually based on preset scripts or simple natural language processing technology to help users solve common problems. 2) Voice recognition and command execution: Allow users to search for functions or get help through voice commands. This method reduces the need for manual input and improves the naturalness of interaction. 3) Personalized recommendation system: Use user historical data, such as transaction records and browsing habits, to provide users with personalized service recommendations. These systems are usually based on machine learning technology and aim to improve the relevance of services and user experience. 4) Image recognition applications: Use image recognition technology to assist users in completing specific tasks, such as simplifying the input process by scanning checks or identity documents.

[0168] Specifically: Intelligent chatbots are limited by preset scripts and simple natural language processing technology, and have limited ability to handle complex queries and specific user needs. Voice recognition and command execution rely on user voice input, which is inconvenient to use in public places and may be affected by factors such as accents and background noise. Personalized recommendation systems usually rely on users' historical data and may not be able to respond to users' current specific needs and situations in a timely manner. Image recognition applications are mainly used to simplify the input process and cannot provide effective help for specific problems and needs encountered by users.

[0169] Therefore, this application effectively makes up for the limitations of existing technologies by comprehensively using multiple technologies and innovative user interaction design, and provides users with a more efficient, intuitive and personalized service experience. This innovation not only enhances the competitiveness of mobile banking APP, but also brings users a more convenient and pleasant user experience.

[0170] Another embodiment of the present application provides an information processing system.

[0171] See also Figure 3 , the information processing system comprises:

[0172] The circle drawing unit 101 is used to determine the target area based on the user's fingertip movement trajectory on the page when the circle drawing matching function is turned on.

[0173] The recognition unit 102 is used to recognize the content in the target area based on the OCR technology and / or the image recognition technology, and analyze the user's context data to determine the user's target data.

[0174] The prediction unit 103 is used to analyze the target data using a prediction model to predict the user's needs or problems.

[0175] The reply unit 104 is used to reply according to the user's needs or questions and display the reply result to the user.

[0176] The specific working process and principle of each of the above modules are detailed in the information processing method provided in the above embodiment, which will not be described one by one here. It depends on the actual situation and is within the protection scope of this application.

[0177] In this embodiment, when the circle matching function of the circle unit 101 is turned on, the target area is determined based on the user's fingertip movement trajectory on the page; the recognition unit 102 recognizes the content in the target area based on OCR technology and / or image recognition technology, and uses the user's context data for analysis to determine the user's target data; the prediction unit 103 uses a prediction model to analyze the target data and predict the user's needs or questions; the reply unit 104 replies according to the user's needs or questions and displays the reply result to the user; that is, the circle function and the matching function are linked; the area that the user needs to answer is determined based on the user's fingertip movement trajectory, and the predicted needs or questions corresponding to the area are replied; the problems caused by the user relying on manual input and page navigation to find information or solve problems are avoided, the efficiency of information processing is improved, and it is intuitive and convenient to meet the user's needs for efficient interaction.

[0178] Another embodiment of the present application provides a storage medium on which a computer program is stored, wherein when the computer program is executed by a processor, an information processing method as described in any one of the above embodiments is implemented.

[0179] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0180] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0181] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0182] Another embodiment of the present invention provides an electronic device, such as Figure 4 As shown, including:

[0183] One or more processors 201.

[0184] The storage device 202 stores one or more programs.

[0185] When one or more programs are executed by one or more processors 201 , the one or more processors 201 implement an information processing method as described in any one of the above embodiments.

[0186] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart.

[0187] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

[0188] Although several specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Certain features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination.

[0189] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0190] The features recorded in the various embodiments in this specification can be replaced or combined with each other, and the same and similar parts between the various embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The system and system embodiments described above are only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative work.

[0191] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0192] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An information processing method, characterized in that: include: When the circle matching function is turned on, the target area is determined based on the user's fingertip movement trajectory on the page; Identify the content in the target area based on OCR technology and / or image recognition technology, and analyze using the user's context data to determine the user's target data; Analyzing the target data using a prediction model to predict the needs or problems of the user; According to the needs or questions of the user, a reply is made and the reply result is displayed to the user.

2. The information processing method according to claim 1, characterized in that: The OCR technology and / or image recognition technology are used to identify the content in the target area, including: If the content in the target area includes text, using the OCR technology to recognize the text in the target area; If the content in the target area includes an image or an interface element, the image recognition technology is used to recognize the image or the interface element in the target area.

3. The information processing method according to claim 1, characterized in that: The step of replying according to the needs or questions of the user and displaying the reply result to the user includes: Searching for the needs or questions of the user to obtain search results corresponding to the needs or questions; The search results corresponding to the requirements or questions are displayed to the user.

4. The information processing method according to claim 1, characterized in that: The step of replying according to the needs or questions of the user and displaying the reply result to the user includes: Transmitting the user's needs or questions to customer service, and having customer service respond to the user's needs or questions; The customer service responds to the request or question and displays it to the user.

5. The information processing method according to claim 1, characterized in that: The context data includes: user behavior history, transaction path, location information and situation awareness data; wherein the situation awareness data is data perceived in combination with the user's current time, date and environmental factors.

6. The information processing method according to claim 1, characterized in that: The process of building and updating the prediction model includes: Collect user behavior data; Standardizing the time series data in the user's behavior data to obtain training samples; The training samples are trained using an LSTM model to learn the time series pattern of user behavior and obtain a prediction model of user behavior; The prediction model is updated regularly based on the prediction results and user behavior data.

7. The information processing method according to claim 1, characterized in that: When the circle matching function is turned on, before determining the target area based on the fingertip movement trajectory of the user on the page, the method further includes: When it is detected that the number and amplitude of shaking of the mobile phone within the preset time meet the preset conditions, it is determined that the circle matching function is turned on.

8. An information processing system, characterized in that: include: The circle unit is used to determine the target area based on the user's fingertip movement trajectory on the page when the circle matching function is turned on; An identification unit, configured to identify the content in the target area based on OCR technology and / or image recognition technology, and analyze the user's context data to determine the user's target data; A prediction unit, configured to analyze the target data using a prediction model to predict the needs or problems of the user; The reply unit is used to reply according to the needs or questions of the user and display the reply result to the user.

9. An electronic device, characterized in that: include: one or more processors; a storage device having one or more programs stored thereon; When one or more programs are executed by one or more processors, the one or more processors are enabled to implement the information processing method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: A computer program is stored thereon, wherein when the computer program is executed by a processor, the information processing method according to any one of claims 1 to 7 is implemented.

Images