Data interaction system based on artificial intelligence algorithm

Abstract

This invention discloses a data interaction system based on artificial intelligence algorithms, specifically relating to the field of data interaction technology. It includes a data acquisition module that collects user input data via sensors, microphones, and cameras. The input data includes speech, images, and text. A multi-band denoising module divides the input speech data into multiple frequency bands and applies an adaptive denoising algorithm to denoise the speech data based on the noise characteristics of each frequency band. Through the collaborative work of multiple modules, this invention achieves accurate data acquisition, denoising, encrypted transmission, intelligent segmentation and routing, behavior analysis and personalized services, as well as system feedback and optimization. This system is characterized by high efficiency, security, and personalization, significantly improving the accuracy of data interaction and user experience.

Description

Technical Field

[0001] This invention relates to the field of data interaction technology, and more specifically, to a data interaction system based on artificial intelligence algorithms. Background Technology

[0002] Currently, with the rapid development of computer science, big data, cloud computing and other technologies, artificial intelligence technology has moved from theory to practice and has achieved remarkable results in various fields. The continuous breakthroughs in technologies such as deep learning, natural language processing and machine learning have provided strong support for the intelligence and efficiency of artificial intelligence systems. The rapid development of these technologies enables artificial intelligence systems to better understand users' intentions and needs, thereby providing more accurate and personalized services.

[0003] In the field of data interaction, with the popularization of the Internet and the widespread application of smart devices, users' demand for data interaction is growing. Traditional data interaction methods often suffer from problems such as low communication efficiency and long response time, which cannot meet users' needs for efficient and intelligent data interaction. Therefore, a more intelligent and efficient data interaction system is needed to meet users' needs.

[0004] Although existing artificial intelligence technologies have achieved some success in the field of data interaction, some limitations still exist:

[0005] 1. In terms of speech recognition, existing technologies are often affected by factors such as noise and accents, resulting in low recognition accuracy.

[0006] 2. In terms of data transmission, existing technologies often lack sufficient security and are vulnerable to hacker attacks and data breaches.

[0007] 3. Existing technologies also have shortcomings in personalized services, and cannot provide accurate services based on users' actual needs.

[0008] In response to the above situation, the present invention provides a data interaction system based on artificial intelligence algorithms. Summary of the Invention

[0009] In order to overcome the above-mentioned deficiencies of the prior art, the present invention provides a data interaction system based on artificial intelligence algorithms to solve the problems mentioned in the background art.

[0010] To achieve the above objectives, the present invention provides the following technical solution: a data interaction system based on artificial intelligence algorithms, including a data acquisition module that collects user input data through sensors, microphones, and camera devices. The input data includes voice, images, and text.

[0011] A multi-band denoising module is used to divide the input speech data into multiple frequency bands and apply an adaptive denoising algorithm to denoise the speech data according to the noise characteristics of each frequency band. The adaptive denoising algorithm includes:

[0012] The formula for converting speech data to the frequency domain using Fast Fourier Transform is as follows:

[0013] ,in, Frequency domain representation, that is, the k-th frequency component of speech data in the frequency domain. Time-domain speech data, i.e., the nth sampling point of the original speech signal; N is the data length, i.e., the total number of sampling points of the speech signal; j is the imaginary unit, satisfying j 2 =-1, Complex exponential functions are used to map time-domain signals to the frequency domain;

[0014] A denoising filter with dynamically adjustable parameters is applied to each frequency band, wherein the parameters are determined based on the frequency band characteristics and noise intensity. The transfer function of the denoising filter can be expressed as:

[0015] ,in, Let be the transfer function of the filter, and z be a variable in the complex plane representing the filter's delay. and The coefficients of the filter determine its frequency response. It is the order of the filter, representing the complexity of the filter;

[0016] The inverse fast Fourier transform converts the denoised data back to the time domain, and the formula is as follows:

[0017] ,in, This represents the nth sampling point reconstructed in the time domain. It is the value of the k-th frequency component in the frequency domain. This formula converts the frequency domain signal back to the time domain to recover the original speech signal.

[0018] The blockchain data encryption and transmission module is used to divide the denoised data into multiple data blocks, encrypt them through the blockchain, attach a unique identifier generated by the SHA-256 hash algorithm to each data block, and transmit them through multiple independent encrypted data channels. The receiving end uses the unique identifier to verify and reassemble the data blocks.

[0019] The data segmentation and dynamic routing module is used to intelligently segment data into multiple data streams according to the type and size of the data, select the best transmission path based on network conditions and data priority, and monitor network conditions and data transmission progress in real time, and dynamically adjust the transmission path and priority.

[0020] The behavior analysis and service module is used to collect and analyze user interaction data, including voice commands and touch operations, and provide personalized services based on the analysis results. These personalized services include intelligent recommendations and voice assistant optimization.

[0021] Preferably, the sensors of the data acquisition module include a temperature sensor, a pressure sensor, and a light sensor. The temperature sensor has a measurement range of -20℃ to 100℃ and an accuracy of ±0.5℃. The pressure sensor has a range of 0 to 1000 kPa and an accuracy of ±1 kPa. The light sensor has a sensitivity of 0.1 lux.

[0022] The microphone has a pickup range of 0.5-10 meters and a sensitivity of -38dBV / Pa. The camera has a resolution of 1080p-4K and a frame rate of 30-60fps.

[0023] Preferably, in the multi-band noise reduction module, the parameter adjustment method of the noise reduction filter includes:

[0024] The signal-to-noise ratio (SNR) for each frequency band is calculated using the following formula: ,in, It is signal power. It is noise power;

[0025] The type and parameters of the denoising filter, including the filter order and cutoff frequency, are determined based on the SNR.

[0026] The parameters of the noise reduction filter are dynamically adjusted to adapt to changes in different frequency bands and noise intensities.

[0027] Preferably, in the blockchain data encryption and transmission module, the data block encryption method includes: encrypting the data block using the ECC asymmetric encryption algorithm, generating a public key and private key pair, broadcasting the public key and data block through the blockchain network, storing the private key at the sending end or in a secure key management service, and the receiving end using the public key to verify the integrity of the data block and using the private key to decrypt the data block.

[0028] Preferably, in the data segmentation and dynamic routing module, the data flow transmission path selection method includes: evaluating the network conditions of multiple data channels, including bandwidth, latency and packet loss rate; selecting the optimal combination of data channels as the transmission path based on data priority and network conditions; continuously monitoring network conditions during transmission; and dynamically adjusting the transmission path as needed.

[0029] Preferably, in the behavior analysis and service module, the user behavior analysis method includes: collecting user interaction data, including voice commands, touch operations and usage time; using machine learning algorithms to classify and cluster the user interaction data; identifying user behavior patterns and preferences; and providing personalized services based on user behavior patterns and preferences, including intelligent recommendations and voice assistant optimization.

[0030] Preferably, it also includes a feedback and optimization module, used to receive user feedback on system performance and service quality, and to optimize and adjust the system's algorithms and parameters based on the feedback results.

[0031] Preferably, it also includes a method for a data interaction system, comprising the following steps:

[0032] S1. Data Acquisition: The data acquisition module collects the user's voice, image, and text input data through sensors, microphones, and camera devices.

[0033] S2. Multi-band denoising: The multi-band denoising module divides the collected speech data into multiple frequency bands, converts it to the frequency domain using fast Fourier transform, applies an adaptive denoising algorithm based on the noise characteristics of each frequency band, processes it through a denoising filter with dynamically adjustable parameters, and then converts it back to the time domain via inverse fast Fourier transform.

[0034] S3. Data Encryption and Transmission: The blockchain data encryption and transmission module divides the denoised data into multiple data blocks, encrypts them using an asymmetric encryption algorithm, and generates public and private key pairs. The public key is transmitted along with the data blocks, and the private key is properly stored. The data blocks are transmitted through the blockchain with a unique identifier attached and through multiple independent encrypted data channels. The receiving end uses the unique identifier to verify and reassemble the data blocks.

[0035] S4. Data Segmentation and Routing: The data segmentation and dynamic routing module intelligently segments data into multiple data streams based on the type and size of the data, evaluates the network conditions of bandwidth, latency and packet loss rate of multiple data channels, selects the best transmission path based on data priority, and monitors the network conditions and data transmission progress in real time during transmission, dynamically adjusting the transmission path and priority.

[0036] S5, Behavior Analysis and Services: The Behavior Analysis and Services module collects and analyzes users' voice commands and touch operation interaction data, uses machine learning algorithms to classify and cluster them to identify users' behavior patterns and preferences, and provides personalized services based on this, such as intelligent recommendations and voice assistant optimization.

[0037] S6. Feedback and Optimization: The feedback and optimization module receives user feedback on system performance and service quality, and optimizes and adjusts the system's algorithms and parameters based on the feedback results to continuously improve the overall system performance and user experience.

[0038] The technical effects and advantages of this invention are as follows:

[0039] By integrating a multi-band noise reduction module and a blockchain data encryption and transmission module, efficient processing and secure transmission of user input data are achieved. The multi-band noise reduction module utilizes an adaptive noise reduction algorithm to accurately identify and remove noise from voice data, improving the accuracy and clarity of speech recognition. Meanwhile, the blockchain data encryption and transmission module ensures the security and integrity of data during transmission by segmenting data blocks, encrypting them, and attaching unique identifiers, effectively preventing data leakage and tampering. Furthermore, the data segmentation and dynamic routing module intelligently selects the optimal transmission path based on network conditions and data priority, further improving the efficiency and reliability of data transmission. Attached Figure Description

[0040] Figure 1 This is a system diagram of the present invention.

[0041] Figure 2 This is a flowchart of the present invention.

[0042] Figure 3 This is a flowchart of the multi-band noise reduction process of the present invention.

[0043] Figure 4 This is a schematic diagram of data segmentation and dynamic routing according to the present invention.

[0044] The attached diagram is labeled as follows: 1. Data acquisition module; 2. Multi-band noise reduction module; 3. Blockchain data encryption and transmission module; 4. Data segmentation and dynamic routing module; 5. Behavior analysis and service module; 6. Feedback and optimization module. Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] This invention proposes a data interaction system based on artificial intelligence algorithms. The implementation scenario can be in an intelligent medical assistant system, where medical staff frequently interact with the system, including voice input of medical records, image upload of examination results, and text input of diagnostic suggestions. The system needs to accurately process data to assist medical decision-making and optimize services, such as ensuring clear voice, secure and efficient data transmission, and providing personalized medical assistance during remote diagnosis. This is the implementation scenario of this invention.

[0047] The system integrates a data acquisition module 1, a multi-band noise reduction module 2, a blockchain data encryption and transmission module 3, a data segmentation and dynamic routing module 4, a behavior analysis and service module 5, and a feedback and optimization module 6. It aims to improve the accuracy and security of data interaction, provide personalized services, and continuously optimize system performance. The specific modules are as follows:

[0048] Data acquisition module 1 collects user input data through sensors, a microphone, and a camera. Input data includes voice, images, and text. The sensors include a temperature sensor, a pressure sensor, and a light sensor. The temperature sensor has a measurement range of -20℃ to 100℃ and an accuracy of ±0.5℃; the pressure sensor has a range of 0-1000kPa and an accuracy of ±1kPa; the light sensor has a sensitivity of 0.1 lux. The microphone has a pickup range of 0.5-10 meters and a sensitivity of -38dBV / Pa; the camera has a resolution of 1080p to 4K and a frame rate of 30-60fps.

[0049] The application environment can be described as follows: In hospital wards, temperature sensors with an accuracy of -20℃ to 100℃ and ±0.5℃ monitor the ambient temperature to prevent overheating and overload; pressure sensors with an accuracy of 0-1000kPa and ±1kPa monitor the pressure of infusion bags to ensure stable infusion; light sensors with a sensitivity of 0.1lux intelligently adjust the light according to the ambient light to protect patients' eye health; microphones with a pickup range of 0.5-10 meters and a sensitivity of -38dBV / Pa accurately collect medical staff's voice commands and patients' symptom descriptions; high-definition cameras with 1080p-4K resolution and 30-60fps capture images of wounds and medical records, and capture medical staff's hand gestures in real time. Multiple devices work together to collect diverse data, injecting intelligent power into the medical process.

[0050] Multi-band denoising module 2 is used to divide the input speech data into multiple frequency bands and apply an adaptive denoising algorithm to denoise the speech data according to the noise characteristics of each frequency band. The adaptive denoising algorithm includes: converting the speech data to the frequency domain through a fast Fourier transform, the formula of which is: ,in, Frequency domain representation, that is, the k-th frequency component of speech data in the frequency domain. Time-domain speech data, i.e., the nth sampling point of the original speech signal; N is the data length, i.e., the total number of sampling points of the speech signal; j is the imaginary unit, satisfying j 2 =-1, Complex exponential functions are used to map time-domain signals to the frequency domain;

[0051] A denoising filter with dynamically adjustable parameters is applied to each frequency band, where the parameters are determined based on the frequency band characteristics and noise intensity. The transfer function of the denoising filter can be expressed as:

[0052] ,in, Let be the transfer function of the filter, and z be a variable in the complex plane representing the filter's delay. and The coefficients of the filter determine its frequency response. It is the order of the filter, representing the complexity of the filter;

[0053] The inverse fast Fourier transform converts the denoised data back to the time domain, and the formula is as follows:

[0054] ,in, This represents the nth sampling point reconstructed in the time domain. It is the value of the k-th frequency component in the frequency domain. This formula converts the frequency domain signal back to the time domain to recover the original speech signal.

[0055] In the multi-band noise reduction module 2, the parameter adjustment methods for the noise reduction filter include:

[0056] The signal-to-noise ratio (SNR) for each frequency band is calculated using the following formula: ,in, It is signal power. It is noise power;

[0057] The type and parameters of the denoising filter, including the filter order and cutoff frequency, are determined based on the SNR.

[0058] The parameters of the noise reduction filter are dynamically adjusted to adapt to changes in different frequency bands and noise intensities.

[0059] The application environment can be described as follows: In a noisy emergency room, voice commands are mixed with equipment alarms and human noise. The system analyzes the noise according to the frequency band characteristics, such as low-frequency humming of equipment (50-200Hz) and high-frequency noise of human voices (1kHz-5kHz). After analyzing the speech spectrum through Fast Fourier Transform, noise reduction filters are set for each frequency band. If strong noise is encountered, the filter parameters are precisely set according to the signal-to-noise ratio (SNR) of a certain frequency band, such as 5dB. A high-order band-stop filter of order 8 is selected, and the cutoff frequency is finely adjusted according to the noise frequency band. Dynamic optimization is performed every 5 milliseconds. After inverse transformation, clear speech is restored. After noise reduction, the speech recognition accuracy is improved by 40%, clearing obstacles for remote consultation and command transmission.

[0060] The blockchain data encryption and transmission module 3 is used to divide the denoised data into multiple data blocks, encrypt them through the blockchain, and attach a unique identifier generated by the SHA-256 hash algorithm to each data block. The data blocks are then transmitted through multiple independent encrypted data channels. The receiving end uses the unique identifier to verify and reassemble the data blocks. In the blockchain data encryption and transmission module 3, the encryption method for the data blocks includes: encrypting the data blocks using the ECC asymmetric encryption algorithm, generating a public key and private key pair, broadcasting the public key and the data blocks through the blockchain network, storing the private key at the sending end or in a secure key management service, and the receiving end using the public key to verify the integrity of the data blocks and using the private key to decrypt the data blocks.

[0061] The application environment can be described as follows: Medical data is related to privacy. After the system segments and denoises the medical records and image data blocks, it uses ECC to encrypt them into public-private key pairs. The public key is broadcast with the data blocks, and the private key is stored on the hospital's encryption server or a professional KMS. Each data block is assigned a unique SHA-256 hash identifier and transmitted to the cloud storage or remote medical terminal through multiple UDP optimized encryption channels. The receiving end verifies the identifier to prevent tampering and decrypts it with the private key, ensuring that the entire process of medical data flow is encrypted, traceable, and leak-proof. The data tampering detection rate exceeds 99.9%, strictly safeguarding the lifeline of medical privacy.

[0062] The data segmentation and dynamic routing module 4 is used to intelligently segment data into multiple data streams according to the type and size of the data, and select the best transmission path based on network conditions and data priority. At the same time, it monitors the network conditions and data transmission progress in real time, and dynamically adjusts the transmission path and priority. In the data segmentation and dynamic routing module 4, the data stream transmission path selection method includes: evaluating the network conditions of multiple data channels, including bandwidth, latency and packet loss rate; selecting the optimal combination of data channels as the transmission path according to data priority and network conditions; continuously monitoring the network conditions during transmission; and dynamically adjusting the transmission path as needed.

[0063] The application environment can be described as follows: medical data is diverse, including medical record texts, images, and real-time physiological monitoring streams. The data size is measured in KB for texts and is intelligently segmented based on image resolution. For example, in live surgery, high-definition images are layered according to frame rate. Network bandwidth, latency, and packet loss rate are evaluated in real time. Surgical control commands prioritize low-latency channels, while images are routed to high-bandwidth, stable paths. When the network fluctuates, such as a sudden 30% drop in bandwidth, intelligent routing switches paths within 10 milliseconds to ensure zero latency for medical commands, stable and accurate data transmission, a 35% increase in bandwidth utilization, and packet loss rate controlled within 1%.

[0064] Behavior Analysis and Service Module 5 is used to collect and analyze user interaction data, including voice commands and touch operations, and provide personalized services based on the analysis results. Personalized services include intelligent recommendations and voice assistant optimization. In Behavior Analysis and Service Module 5, user behavior analysis methods include: collecting user interaction data, including voice commands, touch operations and usage time; using machine learning algorithms to classify and cluster user interaction data; identifying user behavior patterns and preferences; and providing personalized services based on user behavior patterns and preferences, including intelligent recommendations and voice assistant optimization.

[0065] The application environment can be described as follows: collecting medical and nursing operation data, such as voice command semantics (voice recognition accuracy 98%), device touch trajectory (coordinate accuracy 0.05mm), and usage time periods. Deep neural network clustering analysis is used to analyze, for example, a doctor's habit of morning ward rounds and preference for imaging diagnosis. Based on this, the system intelligently recommends similar case references and imaging analysis tools, optimizes the voice assistant interaction logic, proactively pushes ward round tasks in the morning, and increases the volume according to habits to improve medical efficiency and experience. The personalized recommendation accuracy rate reaches 88%.

[0066] Feedback and optimization module 6 is used to receive user feedback on system performance and service quality, and to optimize and adjust the system's algorithms and parameters based on the feedback results.

[0067] The application environment can be described as follows: The system has multiple feedback ports, allowing medical staff to evaluate performance through the interface, provide suggestions in the community, and report anomalies through logs. For example, if a remote consultation encounters a lag, the network adaptation problem can be located through NLP sentiment analysis and clustering. Based on this, the algorithm can be optimized, such as adjusting routing strategy weights, parameters, and iterative functions. Closed-loop optimization is carried out based on performance indicators and satisfaction surveys, with 25% of functions iterated annually, driving the system to adapt to the needs of medical innovation.

[0068] The workflow of this invention is as follows:

[0069] S1. Data acquisition: The data acquisition module 1 acquires the user's voice, image, and text input data through sensors, microphones, and camera devices.

[0070] S2. Multi-band denoising: The multi-band denoising module 2 divides the collected speech data into multiple frequency bands, converts it to the frequency domain using fast Fourier transform, applies an adaptive denoising algorithm based on the noise characteristics of each frequency band, processes it through a denoising filter with dynamically adjustable parameters, and then converts it back to the time domain via inverse fast Fourier transform.

[0071] S3, Data Encryption and Transmission: Blockchain data encryption and transmission module 3 divides the denoised data into multiple data blocks, encrypts them using an asymmetric encryption algorithm, and generates public and private key pairs. The public key is transmitted along with the data blocks, and the private key is properly stored. The data blocks are transmitted through the blockchain with a unique identifier attached and through multiple independent encrypted data channels. The receiving end uses the unique identifier to verify and reassemble the data blocks.

[0072] S4. Data Segmentation and Routing: The data segmentation and dynamic routing module 4 intelligently segments data into multiple data streams based on the type and size of the data, evaluates the network conditions of bandwidth, latency and packet loss rate of multiple data channels, selects the best transmission path based on data priority, and monitors the network conditions and data transmission progress in real time during transmission, dynamically adjusting the transmission path and priority.

[0073] S5, Behavior Analysis and Services: Behavior Analysis and Services module 5 collects and analyzes users' voice commands and touch operation interaction data, uses machine learning algorithms to classify and cluster them to identify users' behavior patterns and preferences, and provides personalized services based on this, such as intelligent recommendations and voice assistant optimization.

[0074] S6. Feedback and Optimization: The feedback and optimization module 6 receives user feedback on system performance and service quality, and optimizes and adjusts the system's algorithms and parameters based on the feedback results to continuously improve the overall system performance and user experience.

[0075] In summary, this embodiment provides a data interaction system based on artificial intelligence algorithms. Through the collaborative work of multiple modules, the system achieves accurate data collection, noise reduction, encrypted transmission, intelligent segmentation and routing, behavior analysis and personalized services, as well as system feedback and optimization. The system is characterized by high efficiency, security, and personalization, and can significantly improve the accuracy of data interaction and user experience.

[0076] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A data interaction system based on artificial intelligence algorithms, characterized in that: Includes a data acquisition module (1), which collects user input data through sensors, microphones and camera devices. The input data includes voice, images and text. A multi-band denoising module (2) is used to divide the input speech data into multiple frequency bands and apply an adaptive denoising algorithm to denoise the speech data according to the noise characteristics of each frequency band. The adaptive denoising algorithm includes: The formula for converting speech data to the frequency domain using Fast Fourier Transform is as follows: ,in, Frequency domain representation, that is, the k-th frequency component of speech data in the frequency domain. Time-domain speech data, i.e., the nth sampling point of the original speech signal; N is the data length, i.e., the total number of sampling points of the speech signal; j is the imaginary unit, satisfying j 2 =-1, Complex exponential functions are used to map time-domain signals to the frequency domain; A denoising filter with dynamically adjustable parameters is applied to each frequency band, wherein the parameters are determined based on the frequency band characteristics and noise intensity. The transfer function of the denoising filter can be expressed as: ,in, Let be the transfer function of the filter, and z be a variable in the complex plane representing the filter's delay. and The coefficients of the filter determine its frequency response. It is the order of the filter, representing the complexity of the filter; The inverse fast Fourier transform converts the denoised data back to the time domain, and the formula is as follows: ,in, This represents the nth sampling point reconstructed in the time domain. It is the value of the k-th frequency component in the frequency domain. This formula converts the frequency domain signal back to the time domain to recover the original speech signal. The blockchain data encryption and transmission module (3) is used to divide the denoised data into multiple data blocks, encrypt them through the blockchain, and attach a unique identifier generated by the SHA-256 hash algorithm to the data blocks. The data blocks are transmitted through multiple independent encrypted data channels, and the receiving end uses the unique identifier to verify and reassemble the data blocks. The data segmentation and dynamic routing module (4) is used to intelligently segment data into multiple data streams according to the type and size of the data, and select the best transmission path based on network conditions and data priority. At the same time, it monitors network conditions and data transmission progress in real time and dynamically adjusts the transmission path and priority. The behavior analysis and service module (5) is used to collect and analyze user interaction data, including voice commands and touch operations, and provide personalized services based on the analysis results. The personalized services include intelligent recommendations and voice assistant optimization.

2. The data interaction system based on artificial intelligence algorithms according to claim 1, characterized in that: The data acquisition module (1) includes a temperature sensor, a pressure sensor and a light sensor. The temperature sensor has a measurement range of -20℃ to 100℃ and an accuracy of ±0.5℃. The pressure sensor has a range of 0 to 1000 kPa and an accuracy of ±1 kPa. The light sensor has a sensitivity of 0.1 lux. The microphone has a pickup range of 0.5-10 meters and a sensitivity of -38dBV / Pa. The camera has a resolution of 1080p-4K and a frame rate of 30-60fps.

3. The data interaction system based on artificial intelligence algorithms according to claim 1, characterized in that: In the multi-band noise reduction module (2), the parameter adjustment method of the noise reduction filter includes: The signal-to-noise ratio (SNR) for each frequency band is calculated using the following formula: ,in, It is signal power. It is noise power; The type and parameters of the denoising filter, including the filter order and cutoff frequency, are determined based on the SNR. The parameters of the noise reduction filter are dynamically adjusted to adapt to changes in different frequency bands and noise intensities.

4. The data interaction system based on artificial intelligence algorithms according to claim 1, characterized in that: In the blockchain data encryption and transmission module (3), the encryption method of the data block includes: encrypting the data block using the ECC asymmetric encryption algorithm, generating a public key and a private key pair, broadcasting the public key and the data block through the blockchain network, storing the private key in the sending end or in a secure key management service, and using the public key to verify the integrity of the data block and using the private key to decrypt the data block.

5. The data interaction system based on artificial intelligence algorithms according to claim 1, characterized in that: In the data segmentation and dynamic routing module (4), the data flow transmission path selection method includes: evaluating the network conditions of multiple data channels, including bandwidth, latency and packet loss rate; selecting the optimal data channel combination as the transmission path according to data priority and network conditions; continuously monitoring the network conditions during transmission; and dynamically adjusting the transmission path as needed.

6. The data interaction system based on artificial intelligence algorithms according to claim 1, characterized in that: In the behavior analysis and service module (5), the user behavior analysis method includes: collecting user interaction data, including voice commands, touch operations and usage time; using machine learning algorithms to classify and cluster user interaction data; identifying user behavior patterns and preferences; and providing personalized services based on user behavior patterns and preferences, including intelligent recommendations and voice assistant optimization.

7. The data interaction system based on artificial intelligence algorithms according to claim 1, characterized in that: It also includes a feedback and optimization module (6), which is used to receive user feedback on system performance and service quality, and optimize and adjust the system's algorithms and parameters based on the feedback results.

8. The data interaction system based on artificial intelligence algorithms according to claim 1, characterized in that: It also includes methods for data interaction systems, comprising the following steps: S1, Data Acquisition: The data acquisition module (1) acquires the user's voice, image and text input data through sensors, microphones and camera devices; S2, Multi-band noise reduction, The multi-band noise reduction module (2) divides the collected speech data into multiple frequency bands, converts it to the frequency domain using fast Fourier transform, applies an adaptive noise reduction algorithm according to the noise characteristics of each frequency band, processes it through a noise reduction filter with dynamically adjustable parameters, and then converts it back to the time domain via inverse fast Fourier transform. S3, Data Encryption and Transmission, Blockchain Data Encryption and Transmission Module (3) divides the denoised data into multiple data blocks, uses an asymmetric encryption algorithm to encrypt and generate public and private key pairs, transmits the public key together with the data blocks, keeps the private key properly, and transmits the data blocks through the blockchain with a unique identifier attached and through multiple independent encrypted data channels. The receiving end uses the unique identifier to verify and reassemble the data blocks. S4. Data segmentation and routing, the data segmentation and dynamic routing module (4) intelligently segments the data into multiple data streams based on the type and size of the data, evaluates the network conditions of the bandwidth, latency and packet loss rate of multiple data channels, selects the best transmission path in combination with data priority, and monitors the network conditions and data transmission progress in real time during the transmission process, and dynamically adjusts the transmission path and priority. S5, Behavior Analysis and Services, Behavior Analysis and Services module (5) collects and analyzes users’ voice commands and touch operation interaction data, uses machine learning algorithms to classify and cluster them to identify users’ behavior patterns and preferences, and provides personalized services based on this, such as intelligent recommendation and voice assistant optimization; S6. Feedback and Optimization: The feedback and optimization module (6) receives user feedback on system performance and service quality, and optimizes and adjusts the system's algorithms and parameters based on the feedback results to continuously improve the overall system performance and user experience.

Images