CN117373036A - Data analysis processing method based on intelligent AI - Google Patents
Data analysis processing method based on intelligent AI Download PDFInfo
- Publication number
- CN117373036A CN117373036A CN202311385108.2A CN202311385108A CN117373036A CN 117373036 A CN117373036 A CN 117373036A CN 202311385108 A CN202311385108 A CN 202311385108A CN 117373036 A CN117373036 A CN 117373036A
- Authority
- CN
- China
- Prior art keywords
- data
- groups
- values
- bandwidth
- different
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007405 data analysis Methods 0.000 title claims abstract description 20
- 238000003672 processing method Methods 0.000 title claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 62
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000012163 sequencing technique Methods 0.000 claims description 24
- 230000005540 biological transmission Effects 0.000 claims description 18
- 238000004458 analytical method Methods 0.000 claims description 15
- 201000010099 disease Diseases 0.000 claims description 15
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 15
- 238000005516 engineering process Methods 0.000 claims description 6
- 238000005192 partition Methods 0.000 claims description 6
- 238000000638 solvent extraction Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 description 6
- 238000012015 optical character recognition Methods 0.000 description 5
- 230000004927 fusion Effects 0.000 description 4
- 238000012165 high-throughput sequencing Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 239000008280 blood Substances 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 238000004393 prognosis Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 208000006011 Stroke Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013079 data visualisation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000012774 diagnostic algorithm Methods 0.000 description 1
- 238000002405 diagnostic procedure Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/19007—Matching; Proximity measures
- G06V30/19093—Proximity measures, i.e. similarity or distance measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/09—Supervised learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/191—Design or setup of recognition systems or techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
- G06V30/19107—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/191—Design or setup of recognition systems or techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
- G06V30/19173—Classification techniques
Abstract
The invention discloses a data analysis processing method based on intelligent AI, which relates to the technical field of PSA data processing and solves the problem that the processing process is not comprehensive because image data with similar characteristics and numerical data cannot be combined and processed.
Description
Technical Field
The invention relates to the technical field of PSA data processing, in particular to a data analysis processing method based on intelligent AI.
Background
In the PSA treatment process, a large amount of PSA characteristic sample data is generated, and in order to confirm the treatment result, a designated AI treatment model is required to be used for treating the characteristic sample data, and the treatment result is displayed for external personnel to check.
The patent publication No. CN115472298B discloses an AI-based intelligent analysis system and method for high-throughput sequencing data, and belongs to the technical field of intelligent analysis of high-throughput sequencing data. The system comprises a high-throughput sequencing data acquisition module, a platform construction module, a flow management module, a personalized data analysis module and a visual output module; the high-throughput sequencing data acquisition module, the platform construction module and the flow management module are sequentially connected; the output end of the flow management module is connected with the input end of the personalized data analysis module; the output end of the personalized data analysis module is connected with the input end of the visual output module. According to the invention, analysis work of mass data can be intelligently processed by utilizing AI, a domestic AI data analysis collaboration platform based on visualization and flow is established, data visualization display of NCS-level research results is realized, and work efficiency of researchers is improved.
In the process of processing the PSA disease data, the image data and the numerical data are in separate states, so that the processing is needed one by one, then the processed results are combined to confirm the corresponding results, in this way, the image data with similar characteristics cannot be combined with the numerical data, so that the processing is not comprehensive, and in the process of adopting an AI model to process the data, the similarity between the data packets is not fully considered, so that the processing progress of the whole processing is affected, and the processing efficiency is slower.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a data analysis processing method based on intelligent AI, which solves the problem that the processing process is not comprehensive due to the fact that image data with similar characteristics and numerical data cannot be combined for processing.
In order to achieve the above purpose, the invention is realized by the following technical scheme: the data analysis processing method based on the intelligent AI comprises the following steps:
s1, recognizing PSA disease feature data in a data sample through a set intelligent AI, wherein the PSA disease feature data comprises a feature image and feature values, the recognized feature image is formulated as a high-dimensional feature, and the recognized feature values are formulated as low-dimensional features;
s2, recognizing high-dimensional data in the high-dimensional features by adopting an OCR image numerical recognition technology, and then carrying out next processing on the recognized high-dimensional data;
s3, extracting corresponding low-dimensional data from the drawn low-dimensional features, analyzing the similarity between a plurality of different high-dimensional data and the low-dimensional data belonging to the same source IP, and locking similar data classification packets, wherein the specific mode is as follows:
s31, distinguishing high-dimensional data and low-dimensional data belonging to the same attribute, partitioning the data belonging to the same attribute, and after partitioning a plurality of different data, confirming a plurality of groups of partitioned data types;
s32, confirming the similarity of the data in each different partition data class, binding the data with the ultrahigh similarity, and locking similar data packets, wherein the specific mode is as follows:
s321, carrying out clustering treatment on different data in the partition data class, locking a group of dots, constructing a clustering circle by the dots, arranging a coordinate system in the clustering circle, and then carrying out distribution sequencing on different high-dimensional data and low-dimensional data according to different parameters in the data to generate a plurality of clustering points corresponding to the periphery of the dots;
s322, confirming a group of clustered circles according to the furthest clustered points and the circular points, confirming the radius value of the clustered circles, and calibrating the radius value as R i Wherein i represents different data;
s323 and then followThe dot starts, a nearest point position is searched, a distance value is determined, then a next nearest point position is searched from the selected point position, a plurality of groups of different distance values are determined, a plurality of groups of distance values are subjected to mean value processing, and a mean value J to be processed is confirmed i Wherein i represents different data;
s324, FX is adopted i ＝R i ×C1+J i Obtaining characteristic value FX of corresponding cluster circle by using XC 2 i Wherein C1 and C2 are both preset fixed coefficient factors;
s325, characteristic values FX of different clustering circles i Sorting from small to large, dividing a plurality of groups of characteristic values with the distance difference not exceeding X1 into the same classification value, and binding corresponding data according to the data corresponding to the characteristic values to obtain similar data packets, wherein X1 is a preset value;
s4, confirming transmission parameters of a plurality of groups of different similar data packets, wherein the transmission parameters comprise bandwidth use parameters and data transmission time interval parameters, randomly selecting two groups of similar data packets according to the transmission parameters for one-to-one analysis, and dividing the two groups of similar data packets into the same-stage processing packets according to analysis results, wherein the specific mode is as follows:
s41, sorting bandwidth use parameters corresponding to similar data packets according to time, sorting different bandwidth values to generate a bandwidth sorting sequence Dk (T1, T2, … …, tn), wherein k represents different similar data packets, and T represents a bandwidth value;
sequencing the data transmission time interval parameters according to time, sequencing different interval parameters to generate an interval parameter sequencing sequence Jk (U1, U2, … …, un), wherein k represents different similar data packets, and U represents an interval parameter;
s42, randomly selecting two groups of similar data packets, confirming a corresponding bandwidth sequencing sequence Dk, determining bandwidth values of the same position in the sequence from the two groups of bandwidth sequencing sequences Dk, performing difference processing on the two groups of bandwidth values, confirming parity bandwidth differences, performing one-to-one processing on the bandwidth values of other same positions, confirming a plurality of groups of parity bandwidth differences, combining the plurality of groups of parity bandwidth differences, and confirming standard bandwidth differences;
then confirming the corresponding interval parameter sequencing sequence Jk, locking interval parameters at the same position in the sequence from the two groups of interval parameter sequencing sequences Jk, performing difference processing on the two groups of interval parameters to confirm the difference value of the parity interval, then performing one-to-one processing on interval parameters at other same positions to confirm a plurality of groups of difference values of the parity interval, combining the plurality of groups of difference values of the parity interval, and determining the standard time difference;
s43, calibrating the confirmed standard bandwidth difference as Bk, calibrating the confirmed standard time difference as Zk, obtaining judging values PDk of two groups of similar data packets by adopting PDk=Bk×A1+Zk×A2, wherein A1 and A2 are preset fixed coefficient factors, and analyzing whether the judging values PDk meet the following conditions: PDk is larger than Y1, wherein Y1 is a preset value, if the PDk is not larger than Y1, the two groups of similar data packets are related, the two groups of similar data packets are divided into the same-stage processing packets, if the PDk is larger than Y1, the two groups of similar data packets are not related, and the two groups of similar data packets are combined with other similar data packets again for analysis;
s5, adopting a preset AI model to process data of two groups of similar data packets of the same-stage processing packet, and directly displaying the processing result for an external operator to check.
Advantageous effects
The invention provides a data analysis processing method based on intelligent AI. Compared with the prior art, the method has the following beneficial effects:
according to the invention, the high-dimensional features and the low-dimensional features of similar features are combined from a large amount of PSA feature data to confirm similar data packets, so that the comprehensiveness of subsequent data processing can be ensured, the high-dimensional features (image data) and the low-dimensional features (numerical data) are fused and associated, and the AI neural network is used as an identification means to find out the connection between different types of features, thereby facilitating the subsequent data processing to achieve a better effect;
confirming transmission data of a plurality of groups of similar data packets obtained through processing, selecting bandwidth values and time interval parameters from the confirmed transmission data, then carrying out combination analysis on the parameters, locking the judging values, then confirming subsequent processing characteristics of different similar data packets according to the determined judging values, combining similar data packets with similar processing characteristics, and then carrying out synchronous processing.
Drawings
FIG. 1 is a schematic flow chart of the method of the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
Referring to fig. 1, the application provides a data analysis and processing method based on intelligent AI, which includes the following steps:
s1, recognizing PSA disease feature data in a data sample through a set intelligent AI, wherein the PSA disease feature data comprises a feature image and a feature value, the recognized feature image is formulated as a high-dimensional feature, the recognized feature value is formulated as a low-dimensional feature, the intelligent AI is specifically a preset AI, and the intelligent AI is constructed by an operator according to experience in advance, and the specific mode of recognition is as follows:
s11, determining sources of different data in a data sample, then, according to specific source IPs, performing IP comparison by an intelligent AI according to a feature library stored in the intelligent AI, and confirming PSA disease feature data, wherein specific source IPs exist in a plurality of source IPs to transmit the PSA disease feature data, so that the PSA disease feature data can be confirmed by locking the corresponding source IPs;
s12, determining feature data belonging to PSA from the feature data of the PSA disease, dividing the numerical data and the image data, then, planning the divided feature image as high-dimensional features and planning the divided feature numerical values as low-dimensional features;
specifically, as the PSA disease characteristic data includes different types of data, but the different types of data have commonality, in order to achieve better data analysis effect, the different types of data need to be combined, and in the combining process, the different types of data need to be preferentially distinguished, and then better combining effect can be achieved;
s2, recognizing high-dimensional data in the high-dimensional features by adopting an OCR image numerical recognition technology, and then performing next processing on the recognized high-dimensional data, wherein the OCR image numerical recognition technology comprises the following steps: by utilizing an optical character recognition technology, character information is extracted from an image, an OCR algorithm can divide a character area in the image and convert the character area into editable text data, so that the recognition mode is the prior art, and redundant description is omitted herein;
s3, extracting corresponding low-dimensional data from the drawn low-dimensional features, analyzing the similarity between a plurality of different high-dimensional data belonging to the same source IP and the low-dimensional data, and locking similar data classification packages, wherein the specific way of analyzing is as follows:
s31, distinguishing high-dimensional data and low-dimensional data belonging to the same attribute, partitioning the data belonging to the same attribute, and after partitioning a plurality of different data, confirming a plurality of groups of partitioned data types;
s32, confirming the similarity of the data in each different partition data class, and binding the data with the ultrahigh similarity so as to lock similar data packets, wherein the specific mode for locking the similar data packets is as follows:
s321, carrying out clustering treatment on different data in the partition data class, locking a group of dots, constructing a clustering circle by the dots, arranging a coordinate system in the clustering circle, and then carrying out distribution sequencing on different high-dimensional data and low-dimensional data according to different parameters in the data to generate a plurality of clustering points corresponding to the periphery of the dots;
s322, confirming a group of clustered circles according to the furthest clustered points and the circular points, confirming the radius value of the clustered circles, and calibrating the radius value as R i Wherein i represents different data, each group of data comprises a plurality of groups of different parameters, each group of parameters has different numerical values, and the data can be ordered according to the different numerical values;
s323, then, starting from the round point, searching the nearest point position, determining the distance value, then, searching the next nearest point position from the selected point position, determining a plurality of groups of different distance values, performing average value processing on the plurality of groups of distance values, and determining the average value J to be processed i Wherein i represents different data;
s324, FX is adopted i ＝R i ×C1+J i Obtaining characteristic value FX of corresponding cluster circle by using XC 2 i Wherein, C1 and C2 are both preset fixed coefficient factors, and the specific value is determined by an operator according to experience;
s325, characteristic values FX of different clustering circles i Sorting from small to large, dividing a plurality of groups of characteristic values with the distance difference not exceeding X1 into the same classification value, binding corresponding data according to the data corresponding to the characteristic values to obtain similar data packets, wherein X1 is a preset value, and the specific value is drawn by an operator according to experience;
specifically, in order to facilitate subsequent data processing, the high-dimensional features and the low-dimensional features of similar features are combined, and similar data packets are confirmed, so that the comprehensiveness of the subsequent data processing can be ensured, the high-dimensional features (image data) and the low-dimensional features (numerical data) are in fusion association, and the AI neural network is used as an identification means to find out the association between different types of features, so that the subsequent data processing can achieve a better effect;
s4, confirming transmission parameters of a plurality of groups of different similar data packets, wherein the transmission parameters comprise bandwidth use parameters and data transmission time interval parameters, two groups of similar data packets are randomly selected according to the transmission parameters to be analyzed one by one, and then the two groups of similar data packets are divided into the same-stage processing packets according to analysis results, wherein the specific mode for carrying out one by one analysis is as follows:
s41, sorting bandwidth use parameters corresponding to similar data packets according to time, sorting different bandwidth values to generate a bandwidth sorting sequence Dk (T1, T2, … …, tn), wherein k represents different similar data packets, and T represents a bandwidth value;
sequencing the data transmission time interval parameters according to time, sequencing different interval parameters to generate an interval parameter sequencing sequence Jk (U1, U2, … …, un), wherein k represents different similar data packets, and U represents an interval parameter;
s42, randomly selecting two groups of similar data packets, confirming a corresponding bandwidth sequencing sequence Dk, determining bandwidth values at the same position in the sequence from the two groups of bandwidth sequencing sequences Dk, performing difference processing on the two groups of bandwidth values, confirming parity bandwidth differences, performing one-to-one processing on the bandwidth values at other same positions, confirming a plurality of groups of parity bandwidth differences, combining the plurality of groups of parity bandwidth differences, confirming standard bandwidth differences, and if ten groups of corresponding bandwidth values exist in one group of sequence and only nine groups of corresponding bandwidth values exist in the other group of sequence, then when processing is performed, the last group of bandwidth values do not participate in processing, and only one group of bandwidth values exist at the same position and do not accord with processing logic;
then confirming the corresponding interval parameter sequencing sequence Jk, locking interval parameters at the same position in the sequence from the two groups of interval parameter sequencing sequences Jk, performing difference processing on the two groups of interval parameters to confirm the difference value of the parity interval, then performing one-to-one processing on interval parameters at other same positions to confirm a plurality of groups of difference values of the parity interval, combining the plurality of groups of difference values of the parity interval, and determining the standard time difference;
s43, calibrating the confirmed standard bandwidth difference as Bk, calibrating the confirmed standard time difference as Zk, and obtaining judging values PDk of two groups of similar data packets by adopting PDk=Bk×A1+Zk×A2, wherein A1 and A2 are preset fixed coefficient factors, the specific values are empirically drawn by operators, and whether the judging values PDk meet the requirement is analyzed: PDk is larger than Y1, wherein Y1 is a preset value, the specific value is drawn by an operator according to experience, if the specific value is satisfied, the two groups of similar data packets are irrelevant, the combination analysis is carried out on the two groups of similar data packets with other similar data packets again, if the specific value is not satisfied, the two groups of similar data packets are relevant, and the two groups of similar data packets are divided into the same-stage processing packets;
specifically, because there are several groups of different similar data packets, if each group of similar data packets is processed one by one through the corresponding AI model, the processing efficiency is relatively slow, and the processing analysis progress of the whole data sample is delayed in this way, so on the premise that the AI model can carry out load, the two groups of similar data packets are synchronously processed, in the processing process, in order to reduce the load, the data classification condition inside the two groups of similar data packets needs to be determined, then, the higher data packets related to the similarity are locked through the data transmission process, and then, the two groups of data packets with higher similarity are synchronously processed, so that the synchronous processing effect is improved, the overall processing efficiency of the data is improved, and the processing time is shortened.
Example two
In the implementation process of the present embodiment, compared with the first embodiment, the present embodiment is mainly directed to a data processing process of a peer-to-peer processing packet, where the processing process is processed by an AI model set originally;
s5, adopting a preset AI model to process data of two groups of similar data packets of the same-stage processing packet, directly displaying the processing result for an external operator to check, accelerating a convergence function of the data by modifying a mature model and dividing data of a new task through a current forefront meta-learning technology and giving AI autonomous learning mode so as to realize modeling of small sample data;
different source data are integrated in a multi-modal fusion manner by retrospectively collecting clinical examination multi-set of clinical characterization data including image data (MRI, CT, etc.), laboratory examination data (blood biochemistry, blood routine, etc.), and clinical characterization data for a large sample volume of stroke patients. On the basis, the supervision type learning of the PSA is carried out on the study object according to the marks of clinical diagnostic specialists, and the corresponding clinical characteristics are learned by the characteristic pyramid neural network framework, so that the mutual combination between low-level characteristics (such as digital characteristics) and high-level characteristics (such as image characteristics) is realized, and the PSA AI auxiliary diagnostic algorithm with high sensitivity and high accuracy is obtained. To improve the clinical acceptability of the algorithm, we introduce attention-back techniques into the model to explain the main judgment basis in PSA diagnostic process;
aiming at the characteristics that the clinical PSA treatment patients are fewer and cannot utilize big data modeling, the subject group adopts the method of expanding the detection types and increasing the detection modes on one hand and disassembles and reforms the existing PSA diagnosis model on the other hand. Under the condition of retaining image recognition and multi-mode fusion algorithm layers, the model is subjected to meta-learning by a transducer algorithm through a strategy of dividing PSA prognosis data, so that the model obtains autonomous learning capacity and higher data convergence speed, and the construction of a PSA high-accuracy and high-sensitivity AI auxiliary prognosis evaluation model under the condition of extremely small sample size is realized. Because the data segmentation and the modal fusion are adopted at the same time, the attention back mark probability is low. In order to ensure clinical applicability, research teams are about to adopt a multi-center verification mode to carry out model verification work in different domestic cooperative units, and properly adjust parameters to optimize the universality of the PSA prevention evaluation model.
Some of the data in the above formulas are numerical calculated by removing their dimensionality, and the contents not described in detail in the present specification are all well known in the prior art.
The above embodiments are only for illustrating the technical method of the present invention and not for limiting the same, and it should be understood by those skilled in the art that the technical method of the present invention may be modified or substituted without departing from the spirit and scope of the technical method of the present invention.
Claims (7)
1. The data analysis processing method based on the intelligent AI is characterized by comprising the following steps:
s1, recognizing PSA disease feature data in a data sample through a set intelligent AI, wherein the PSA disease feature data comprises a feature image and feature values, the recognized feature image is formulated as a high-dimensional feature, and the recognized feature values are formulated as low-dimensional features;
s2, recognizing high-dimensional data in the high-dimensional features by adopting an OCR image numerical recognition technology, and then carrying out next processing on the recognized high-dimensional data;
s3, extracting corresponding low-dimensional data from the drawn low-dimensional features, analyzing the similarity between a plurality of different high-dimensional data belonging to the same source IP and the low-dimensional data, and locking similar data classification packets;
s4, confirming transmission parameters of a plurality of groups of different similar data packets, wherein the transmission parameters comprise bandwidth use parameters and data transmission time interval parameters, randomly selecting two groups of similar data packets according to the transmission parameters for one-to-one analysis, and dividing the two groups of similar data packets into the same-stage processing packets according to analysis results.
2. The data analysis processing method based on intelligent AI according to claim 1, wherein in step S1, the specific way of identifying PSA disease feature data is as follows:
s11, determining sources of different data in a data sample, and then, according to a specific source IP, performing IP comparison by an intelligent AI according to an internally stored feature library to confirm PSA disease feature data;
s12, determining the characteristic data belonging to the PSA from the PSA disease characteristic data, dividing the numerical data and the image data, then, planning the divided characteristic image as a high-dimensional characteristic, and planning the divided characteristic numerical value as a low-dimensional characteristic.
3. The data analysis processing method based on intelligent AI according to claim 1, wherein in step S3, the specific way of analyzing the similarity between the different high-dimensional data and the low-dimensional data is as follows:
s31, distinguishing high-dimensional data and low-dimensional data belonging to the same attribute, partitioning the data belonging to the same attribute, and after partitioning a plurality of different data, confirming a plurality of groups of partitioned data types;
s32, confirming the similarity of the data in each different partition data class, and binding the data with the ultrahigh similarity, so that similar data packets are locked.
4. The data analysis processing method based on intelligent AI of claim 3, wherein in step S32, the specific way of locking the similar data packet is:
s321, carrying out clustering treatment on different data in the partition data class, locking a group of dots, constructing a clustering circle by the dots, arranging a coordinate system in the clustering circle, and then carrying out distribution sequencing on different high-dimensional data and low-dimensional data according to different parameters in the data to generate a plurality of clustering points corresponding to the periphery of the dots;
s322, confirming a group of clustered circles according to the furthest clustered points and the circular points, confirming the radius value of the clustered circles, and calibrating the radius value as R i Wherein i represents different data;
s323, starting from the round point, searching the nearest point position, determining the distance value, searching the next nearest point position from the selected point position, determining a plurality of groups of different distance values, performing average value processing on the plurality of groups of distance values, and determining the average value J to be processed i Wherein i represents different data;
s324, FX is adopted i ＝R i ×C1+J i Obtaining characteristic value FX of corresponding cluster circle by using XC 2 i Wherein C1 and C2 are both preset fixed coefficient factors;
s325, characteristic values FX of different clustering circles i Ordered in a small to large manner, followed by a number of differences no more than X1 apartThe group characteristic values are divided into the same classification values, and then the corresponding data are bundled according to the data corresponding to the characteristic values to obtain similar data packets, wherein X1 is a preset value.
5. The data analysis processing method based on intelligent AI according to claim 1, wherein in step S4, a specific manner of randomly selecting two groups of similar data packets for one-to-one analysis is as follows:
s41, sorting bandwidth use parameters corresponding to similar data packets according to time, sorting different bandwidth values to generate a bandwidth sorting sequence Dk (T1, T2, … …, tn), wherein k represents different similar data packets, and T represents a bandwidth value;
sequencing the data transmission time interval parameters according to time, sequencing different interval parameters to generate an interval parameter sequencing sequence Jk (U1, U2, … …, un), wherein k represents different similar data packets, and U represents an interval parameter;
s42, randomly selecting two groups of similar data packets, confirming a corresponding bandwidth sequencing sequence Dk, determining bandwidth values of the same position in the sequence from the two groups of bandwidth sequencing sequences Dk, performing difference processing on the two groups of bandwidth values, confirming parity bandwidth differences, performing one-to-one processing on the bandwidth values of other same positions, confirming a plurality of groups of parity bandwidth differences, combining the plurality of groups of parity bandwidth differences, and confirming standard bandwidth differences;
then confirming the corresponding interval parameter sequencing sequence Jk, locking interval parameters at the same position in the sequence from the two groups of interval parameter sequencing sequences Jk, performing difference processing on the two groups of interval parameters to confirm the difference value of the parity interval, then performing one-to-one processing on interval parameters at other same positions to confirm a plurality of groups of difference values of the parity interval, combining the plurality of groups of difference values of the parity interval, and determining the standard time difference;
s43, calibrating the confirmed standard bandwidth difference as Bk, calibrating the confirmed standard time difference as Zk, obtaining judging values PDk of two groups of similar data packets by adopting PDk=Bk×A1+Zk×A2, wherein A1 and A2 are preset fixed coefficient factors, and analyzing whether the judging values PDk meet the following conditions: PDk is larger than Y1, wherein Y1 is a preset value, if not, the two groups of similar data packets are related, and the two groups of similar data packets are divided into the same-stage processing packets.
6. The data analysis processing method based on intelligent AI of claim 5, wherein in step S43, if PDk > Y1 is satisfied, two similar data packets are not related, and the combination analysis is performed again with other similar data packets.
7. The intelligent AI-based data analysis processing method of claim 1, further comprising the steps of:
s5, adopting a preset AI model to process data of two groups of similar data packets of the same-stage processing packet, and directly displaying the processing result for an external operator to check.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311385108.2A CN117373036A (en) | 2023-10-24 | 2023-10-24 | Data analysis processing method based on intelligent AI |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311385108.2A CN117373036A (en) | 2023-10-24 | 2023-10-24 | Data analysis processing method based on intelligent AI |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117373036A true CN117373036A (en) | 2024-01-09 |
Family
ID=89390641
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311385108.2A Pending CN117373036A (en) | 2023-10-24 | 2023-10-24 | Data analysis processing method based on intelligent AI |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117373036A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018097765A1 (en) * | 2016-11-24 | 2018-05-31 | Виктор Николаевич ЛИТУЕВ | Method for diagnosing non-infectious diseases on the basis of statistical data processing methods |
| WO2020250995A1 (en) * | 2019-06-11 | 2020-12-17 | 学校法人東海大学 | Morbidity determination assistance device, morbidity determination assistance method, and morbidity determination assistance program |
| CN113129309A (en) * | 2021-03-04 | 2021-07-16 | 同济大学 | Medical image semi-supervised segmentation system based on object context consistency constraint |
| CN115130373A (en) * | 2022-06-08 | 2022-09-30 | 苏州睿远智能科技有限公司 | Intelligent proofreading system and method based on image processing |
| CN115472298A (en) * | 2022-10-28 | 2022-12-13 | 方寸慧医(江苏)生物科技有限公司 | AI-based high-throughput sequencing data intelligent analysis system and method |
-
2023
- 2023-10-24 CN CN202311385108.2A patent/CN117373036A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018097765A1 (en) * | 2016-11-24 | 2018-05-31 | Виктор Николаевич ЛИТУЕВ | Method for diagnosing non-infectious diseases on the basis of statistical data processing methods |
| WO2020250995A1 (en) * | 2019-06-11 | 2020-12-17 | 学校法人東海大学 | Morbidity determination assistance device, morbidity determination assistance method, and morbidity determination assistance program |
| CN113129309A (en) * | 2021-03-04 | 2021-07-16 | 同济大学 | Medical image semi-supervised segmentation system based on object context consistency constraint |
| CN115130373A (en) * | 2022-06-08 | 2022-09-30 | 苏州睿远智能科技有限公司 | Intelligent proofreading system and method based on image processing |
| CN115472298A (en) * | 2022-10-28 | 2022-12-13 | 方寸慧医(江苏)生物科技有限公司 | AI-based high-throughput sequencing data intelligent analysis system and method |
Non-Patent Citations (2)
| Title |
|---|
| 汤建儿;郑祥义;王潇;谢立平;王荣江;陈煜;高建国;: "人工智能多参数MRI在前列腺癌早期诊断中的应用", 中华男科学杂志, no. 09, 20 September 2020 (2020-09-20) * |
| 秦方普;张爱武;王书民;孟宪刚;胡少兴;孙卫东;: "基于谱聚类与类间可分性因子的高光谱波段选择", 光谱学与光谱分析, no. 05, 15 May 2015 (2015-05-15) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113256636B (en) | Bottom-up parasite species development stage and image pixel classification method | |
| CN110322356A (en) | The medical insurance method for detecting abnormality and system of dynamic multi-mode are excavated based on HIN | |
| CN103886030B (en) | Cost-sensitive decision-making tree based physical information fusion system data classification method | |
| CN106202353A (en) | A kind of visable representation method of time series data | |
| CN115359873B (en) | Control method for operation quality | |
| CN102591898B (en) | Bilingual information retrieval apparatus, translation apparatus, and computer readable medium | |
| CN116959725A (en) | Disease risk prediction method based on multi-mode data fusion | |
| CN113160974A (en) | Mental disease biological type mining method based on hypergraph clustering | |
| CN114897085A (en) | Clustering method based on closed subgraph link prediction and computer equipment | |
| CN111524140A (en) | Medical image semantic segmentation method based on CNN and random forest method | |
| CN114358202A (en) | Information pushing method and device based on drug molecule image classification | |
| CN114005530A (en) | Intelligent reminding and monitoring method and system for medical repeated examination and inspection in area | |
| CN109389170A (en) | A kind of gradation type operating condition method for early warning based on 3D convolutional neural networks | |
| CN111370055B (en) | Intron retention prediction model establishment method and prediction method thereof | |
| CN117373036A (en) | Data analysis processing method based on intelligent AI | |
| CN116129182A (en) | Multi-dimensional medical image classification method based on knowledge distillation and neighbor classification | |
| CN113706459B (en) | Detection and simulation repair device for abnormal brain area of autism patient | |
| CN110164504A (en) | Processing method, device and the electronic equipment of two generation sequencing datas | |
| CN116304252A (en) | Communication network fraud prevention method based on graph structure clustering | |
| CN112435230B (en) | Deep learning-based data set generation method and system | |
| CN114496196A (en) | Automatic auditing system for clinical biochemical inspection in medical laboratory | |
| CN114171206A (en) | Model training method, sensing disease prediction method, device, equipment and storage medium | |
| CN112348048A (en) | Multi-structure multi-task deep neural network and training and classifying method thereof | |
| WO2015029158A1 (en) | Data conversion device, data conversion method, and data conversion program | |
| EP3138033B1 (en) | Method and apparatus for performing block retrieval on block to be processed of urine sediment image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |