CN116886104A - A smart medical data analysis method based on artificial intelligence - Google Patents

Abstract

The invention relates to the technical field of data processing, in particular to an intelligent medical data analysis method based on artificial intelligence, which comprises the following steps: acquiring historical intelligent medical data, character types of newly-added intelligent medical data and newly-added intelligent medical data; acquiring an initial frequency contribution weight of each character position of the historical intelligent medical data according to the data length of the historical intelligent medical data; obtaining final weighting frequency of the character type of the historical intelligent medical data according to the adjustment factor of the character type of the newly added intelligent medical data and the weighting frequency of the character type; performing self-adaptive adjustment on the weighted frequencies of other types of characters of the historical intelligent medical data according to the final weighted frequencies of the newly added intelligent medical data type characters of the historical intelligent medical data; and compressing the newly added intelligent medical data according to the adjusted weighting frequency of each type of character. The invention makes the average code length of characters with higher local frequency shorter, and increases the arithmetic coding compression rate.

Description

A smart medical data analysis method based on artificial intelligence

Technical field

The present invention relates to the field of data processing technology, and specifically relates to a smart medical data analysis method based on artificial intelligence.

Background technique

Smart medical data analysis methods have been applied in many fields. For example, in clinical medicine, smart medical data analysis methods can be used to assist doctors in disease diagnosis and treatment decisions. By analyzing and mining large amounts of patient data, potential patterns and patterns of diseases can be discovered, and personalized diagnosis and treatment plans can be provided. Smart medical data has involved all aspects of patients, which has led to the current smart medical treatment having a huge amount of data, which will cause huge resource consumption in terms of storage and transmission. Therefore, an analysis of smart medical data is needed. Method to achieve data compression and reduce storage and transmission costs.

Traditional arithmetic coding can compress data to the ideal shortest length. Arithmetic coding can only compress data with known data frequency. Smart medical data often requires real-time data updates and cannot obtain the optimal fixed length of character intervals. As a result, the ideal compression ratio cannot be achieved.

Contents of the invention

The present invention provides a smart medical data analysis method based on artificial intelligence to solve existing problems.

A smart medical data analysis method based on artificial intelligence of the present invention adopts the following technical solution:

One embodiment of the present invention provides a smart medical data analysis method based on artificial intelligence. The method includes the following steps:

Get historical data, new data, new data type characters and new data;

The initial frequency contribution weight of each character position in the historical data is obtained according to the data length of the historical data;

Obtain the frequency of appearance of new data in historical data; obtain each character in historical data based on the character position of new data in historical data, the initial frequency contribution weight of the character position and the frequency of appearance of new data in historical data. The final frequency contribution weight of the position; the weighted frequency of the new data type character in the historical data is obtained based on the final frequency contribution weight of each character position in the historical data and the new data type character; the weighted frequency of the new data type character in the historical data is obtained according to the appearance of the new data in the historical data The adjustment factor of the new data type characters of historical data is obtained; according to the weighted frequency of the new data type characters of historical data and the adjustment factor of the new data type characters of historical data, the adjustment factor of the new data type characters of historical data is obtained final weighted frequency;

The normalized coefficient is obtained based on the final weighted frequency of the new data character type in historical data; the adjusted weighted frequency of other types of characters in historical data is obtained based on the normalized coefficient; the final weighted frequency sum of the new data character type in historical data is obtained The adjusted weighted frequency of other types of characters in historical data is obtained by adjusting the weighted frequency of all types of characters in historical data; the newly added data is compressed based on the adjusted weighted frequency of all types of characters in historical data.

Preferably, the specific steps of obtaining historical data, new data, new data type characters and new data include the following:

Collect smart medical data through medical equipment to obtain historical smart medical data; after acquiring historical smart medical data, new smart medical data will be added at different times, so it and historical smart medical data should be stored at the same time to obtain new The added smart medical data; record the collected historical smart medical data as historical data, obtain the character type of the newly added smart medical data, record it as the new data type character, and record the newly added smart medical data as new the subsequent data.

Preferably, obtaining the initial frequency contribution weight of each character position in the historical data according to the data length of the historical data includes the following specific steps:

According to the Poisson equation and the data length of historical data, the calculation expression of the initial frequency contribution weight of each character position in the historical data is:

In the formula, Indicates the number/> in historical data Initial frequency contribution weight of character positions;/> Represents the data length of historical data; λ is the Poisson parameter;/> () represents an exponential function with a natural constant as the base.

Preferably, the final frequency contribution weight of each character position in the historical data is obtained based on the character position of the new data appearing in the historical data, the initial frequency contribution weight of the character position and the frequency of appearance of the new data in the historical data. value, including the following specific steps:

For the i-th character position in historical data, the calculation expression to obtain the final frequency contribution weight of the i-th character position in historical data based on the zero-one variable of the i-th character position and the initial frequency contribution weight is:

In the formula, Indicates the number/> in historical data The final frequency contribution weight of character positions;/> Indicates the number/> in historical data Zero-one variable at character position;/> Indicates the number/> in historical data Initial frequency contribution weight of character positions;/> Represents the data length of historical data;

The method for obtaining the zero-one variable at the i-th character position is: for any character of new data, map the character of the new data to the i-th character position in the historical data. If the character at the i-th character position If the character is consistent with the character of the new data, then the zero-one variable V at the i-th character position in the historical data is assigned a value of 1; if the character at the i-th character position is inconsistent with the character of the new data, then the i-th character position in the historical data is inconsistent The zero-one variable V at character position is assigned a value of 0.

Preferably, obtaining the weighted frequency of the new data type characters of the historical data based on the final frequency contribution weight of each character position in the historical data and the new data type characters includes the following specific steps:

For new data character types, the calculation expression to obtain the weighted frequency of this type of character in historical data based on the frequency of occurrence of this type of character in historical data, character position, and the final frequency contribution weight of this character position is:

In the formula, Represents the weighted frequency of new data character types for historical data;/> Represents the historical data The final frequency contribution weight of character positions;/> Represents the frequency of new data character types in historical data in historical data;/> Represents the number of new data character types in historical data in historical data;/> Indicates the data length of historical data.

Preferably, the specific steps of obtaining the adjustment factor of the new data type characters of the historical data based on the number of new data appearing in the historical data are as follows:

For new data type characters, based on the number of occurrences of this type of character in historical data, the calculation expression for obtaining the adjustment factor of this type of character is:

In the formula, Represents the adjustment factor of the new data character type of historical data;/> Represents the number of new data character types in historical data in historical data;/> Indicates the data length of historical data;/> The number of character types representing historical data;/> () represents an exponential function with a natural constant as the base.

Preferably, the final weighted frequency of the new data character type of historical data is obtained based on the weighted frequency of the new data type characters of historical data and the adjustment factor of the new data type characters of historical data. The specific steps include the following:

The product of the adjustment factor of the new data character type of historical data and the weighted frequency of the new data character type of historical data is used as the final weighted frequency of the new data character type of historical data.

Preferably, the specific formula for obtaining the normalization coefficient based on the final weighted frequency of the new data character type of historical data is as follows:

In the formula, Represents the normalization coefficient;/> Represents the final weighted frequency of new data character types for historical data;/> Represents the frequency of new data character types in historical data in historical data.

Preferably, obtaining the adjusted weighted frequencies of other types of characters in historical data based on the normalization coefficient includes the following specific steps:

The product of the initial frequency of characters of other character types in historical data and the normalization coefficient is used as the adjusted weighted frequency of other types of characters in historical data.

Preferably, the newly added data is compressed according to the adjusted weighted frequency of all types of characters in historical data, and the specific steps include the following:

According to the adjusted weighted frequency of each type of character in the historical data, the frequency intervals of all types of characters in the historical data are arranged in descending order according to the adjusted weighted frequency. If there are weighted frequencies of the same size, they are arranged according to their ASCII values in the dictionary. Sort; obtain the frequency intervals corresponding to all types of characters in historical data, and use arithmetic coding to compress the newly added data.

The beneficial effects of the technical solution of the present invention are: traditional arithmetic coding can only compress data with known data frequency, while smart medical data often requires real-time data updates and cannot obtain the optimal fixed length of character intervals. This leads to the problem that the ideal compression rate cannot be achieved. The present invention dynamically adjusts the frequency interval size of each character in arithmetic coding. After each character is read and encoded, the frequency of the characters at each position is adjusted according to the historical distribution of the character. The contribution is weighted to obtain the weighted character interval size, which makes the average code length of characters with higher local frequency shorter and increases the arithmetic coding compression rate.

Description of the drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without exerting creative efforts.

Figure 1 is a step flow chart of a smart medical data analysis method based on artificial intelligence according to the present invention.

Detailed ways

In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended inventive purpose, the following is a detailed implementation of a smart medical data analysis method based on artificial intelligence proposed by the present invention in conjunction with the drawings and preferred embodiments. The method, structure, characteristics and functions are described in detail below. In the following description, different terms "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Additionally, the specific features, structures, or characteristics of one or more embodiments may be combined in any suitable combination.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which the invention belongs.

The specific solution of a smart medical data analysis method based on artificial intelligence provided by the present invention will be described in detail below with reference to the accompanying drawings.

Please refer to Figure 1, which shows a flow chart of a smart medical data analysis method based on artificial intelligence provided by one embodiment of the present invention. The method includes the following steps:

Step S001: Obtain historical smart medical data, new smart medical data character types, and new smart medical data.

It should be noted that smart medical data refers to patient status that needs to be recorded by mobile phones, mainly including patient information, medical records, physiological parameter monitoring data, exercise and activity data, health indicator assessment, health habits and lifestyles, and mental health data. These data include, but are not limited to, blood pressure, heart rate, blood oxygen saturation, respiratory rate, body temperature, blood sugar levels, electroencephalogram data, electrocardiogram data, etc. Collecting this data can help doctors understand the patient's physical condition in real time, thereby ensuring the effectiveness of treatments.

Specifically, smart medical data is collected through relevant medical equipment. The collected smart medical data includes the patient's blood pressure, heart rate, blood oxygen saturation, respiratory rate, body temperature, blood sugar level, electroencephalogram data and electrocardiogram data. The collected smart medical data The data types of medical data are mainly numbers and English symbols. This embodiment does not need to consider the operational inconvenience caused by special data types.

Smart medical data records the patient's physical condition. Data such as blood pressure and blood sugar need to record changes within a certain period of time, and the patient's recovery status can be obtained by analyzing the data change trend. Therefore, after obtaining the historical smart medical data, if new smart medical data will be added at different times in the future, it should be stored together with the historical smart medical data to obtain the new smart medical data; the collected historical wisdom The medical data is recorded as historical data, the character type of the newly added smart medical data is obtained and recorded as the new data type character, and the newly added smart medical data is recorded as the newly added data.

At this point, historical data, new data type characters and new data are obtained.

Step S002: Obtain the initial frequency contribution weight of each character position of the historical smart medical data according to the data length of the historical smart medical data.

It should be noted that arithmetic coding is a coding method that sets a fixed-length frequency interval according to the frequency of each character in the data. The corresponding frequency interval is allocated to the character based on its frequency to achieve the encoding length closest to the data information entropy. . After a data compression, smart medical data needs to continue to be added, and the new data and historical data belong to the same patient measured at different times and need to be classified and stored according to the patient. After the data is updated, if the historical data is decompressed and recompressed, a large amount of calculations will be required, and if the compression is performed at the original frequency, the final compression rate will be insufficient. Therefore, after each character of new data is read in, the local frequency of the character in the historical data is analyzed, the frequency is weighted to obtain the weighted frequency interval, and the character frequency interval is dynamically adjusted to make the character frequency interval correspond to The encoding length is closer to the data information entropy to achieve the ideal compression rate.

It should be further explained that for any character in the historical data, the initial frequency contribution weight of the character position can be obtained according to the position of the character in the historical data; if the character appears more frequently in the historical data, it is speculated that it will be used next time. The new data that comes down has a higher probability of occurrence and can be given a higher weight. If the character appears at the front of the historical data, it means that the probability of the character appearing in the new data is low, and a lower weight will be assigned to it.

Since the patient's physical characteristics do not mutate in a short period of time, but fluctuate within a small range, therefore, for new data, the historical data in adjacent short periods of time is equal to the frequency of the new data and obeys the Poisson curve trend. ; Therefore, the law of its weighting value conforms to the trend of the Poisson curve, so the Poisson curve is used to calculate the initial frequency contribution weight of each character position in each historical data. The Poisson equation parameters are obtained based on the position where each character appears in the historical data and the data length of the historical data, and then the initial frequency contribution weight of each character position in the historical data is obtained based on the Poisson curve trend.

Specifically, based on the Poisson equation and the data length of historical data, the calculation expression for the initial frequency contribution weight of each character position in the historical data is:

In the formula, Indicates the number/> in historical data Initial frequency contribution weight of character positions;/> Represents the data length of historical data; λ is the Poisson parameter;/> () represents an exponential function with a natural constant as the base.

Step S003: Obtain the final frequency contribution weight of each character position of the historical smart medical data and the adjustment factor of the character type of the newly added smart medical data; according to the adjustment factor of the character type of the newly added smart medical data and the character of the historical smart medical data The weighted frequency of the type obtains the final weighted frequency of the character type in the historical wisdom medical data.

1. Obtain the final frequency contribution weight of each character position in historical data.

It should be noted that for the ordinary character frequency calculation method, the contribution of the character occurrence to the frequency at each position of the data is the same, which is the reciprocal of the data length. The contribution of the character is also the total frequency of the character appearing in the data. The countdown. In fact, the local frequency of a character is often more representative of the occurrence probability of the character within a certain period of time. Therefore, the weighted frequency of the character needs to be obtained by weighting the contribution of the character's position on the data to the frequency of the character. Every time a new type of character is added to the data, the local frequencies of other types of characters will be slightly weakened. Therefore, the weighted frequency of the newly added type of characters and the original frequencies of other types of characters are normalized to adjust the entire character frequency of the data. the goal of.

Specifically, for any character of new data, the character of the new data is mapped to the i-th character position in the historical data. If the character at the i-th character position is consistent with the character of the new data, then the character in the historical data The zero-one variable V at the i-th character position is assigned a value of 1; if the character at the i-th character position is inconsistent with the characters of the new data, the zero-one variable V at the i-th character position in the historical data is assigned a value of 0; then Get the zero-one variables of all character positions in the historical data.

For the i-th character position in historical data, the calculation expression to obtain the final frequency contribution weight of the i-th character position in historical data based on the zero-one variable of the character position and the initial frequency contribution weight is:

In the formula, Indicates the number/> in historical data The final frequency contribution weight of character positions;/> Indicates the number/> in historical data Zero-one variable at character position;/> Indicates the number/> in historical data Initial frequency contribution weight of character positions;/> Indicates the data length of historical data.

At this point, the final frequency contribution weight of each character position in the historical data is obtained.

2. Obtain the weighted frequency and adjustment factor of the new data type characters of historical data.

It should be noted that the probability that the new data is a character of this type is predicted based on the character position of each type of character in historical data. The probability that the new data is a character of this type depends more on the historical data of this type of character. If the local frequency of this type of character in historical data is large, it can mean that the probability of new data being this type of character is greater, and its frequency range will also be increased; if the local frequency of this type of character in historical data is If the local frequency is low and the probability of appearing at the front character position in historical data is high, it means that the character has left the character aggregation interval with a higher probability of appearing in historical data. It means that the probability of new data being this type of character is also low. It can be Appropriately lower its frequency range in historical data.

Specifically, for the new data character type, the calculation expression to obtain the weighted frequency of this type of character in historical data based on the frequency of occurrence of this type of character in historical data, character position, and the final frequency contribution weight of this character position is:

In the formula, Represents the weighted frequency of new data character types for historical data;/> Represents the historical data The final frequency contribution weight of character positions;/> Represents the frequency of new data character types in historical data in historical data;/> Represents the number of new data character types in historical data in historical data;/> Indicates the data length of historical data.

It should be noted that the weight of the newly added data type characters should be adjusted considering the frequency of occurrence in historical data. For characters with a small overall frequency of occurrence in historical data, the probability of being new data is also small, so they should be The frequency range should be adjusted smaller relative to the existing weighted frequency; for characters that appear more frequently in historical data, the probability of new data is also high, so the frequency range should be adjusted smaller relative to the existing weighted frequency. Larger; therefore it is necessary to obtain the adjustment factor for the new data type character of the historical data.

Specifically, for new data type characters, based on the number of occurrences of this type of character in historical data, the calculation expression for obtaining the adjustment factor of this type of character is:

In the formula, Represents the adjustment factor of the new data character type of historical data;/> Represents the number of new data character types in historical data in historical data;/> Indicates the data length of historical data;/> The number of character types representing historical data;/> () represents an exponential function with a natural constant as the base.

At this point, the weighted frequency and adjustment factor of the new data type characters of historical data are obtained.

3. Obtain the final weighted frequency of the new data type characters of historical data.

The calculation expression to obtain the final weighted frequency of each type of character in historical data based on the weighted frequency and adjustment factor of the new data type characters in historical data is:

In the formula, Represents the final weighted frequency of new data character types for historical data;/> Represents the adjustment factor of the new data character type of historical data;/> Represents the weighted frequency of new data character types for historical data.

At this point, the final weighted frequency of the new data type characters of historical data is obtained.

Step S004: Adaptively adjust the weighted frequency of other types of characters in the historical smart medical data according to the final weighted frequency of the new smart medical data type characters in the historical smart medical data; adjust the weighted frequency of each type of character according to the historical smart medical data Compress the newly added smart medical data according to the frequency.

It should be noted that due to the new data, the character frequencies of the new data types in the historical data change, while the character frequencies of other types of historical data do not change. Therefore, the character frequencies of all types of characters in the historical data at this time are and may not be 1. This is because it is necessary to normalize the character frequencies of other types of characters in the historical data while ensuring that the character frequencies of the new data types in the historical data remain unchanged with the final weighted frequency obtained by calculation, so that the frequencies of other types of characters can be obtained. Frequency interval size.

Specifically, the calculation expression of the normalization coefficient is obtained based on the final weighted frequency of the new data character type in the historical data:

In the formula, Represents the normalization coefficient;/> Represents the final weighted frequency of new data character types for historical data;/> Represents the frequency of new data character types in historical data in historical data.

The calculation expression for the adjusted weighted frequency of other types of characters in historical data is obtained based on the normalization coefficient:

In the formula, Indicates the adjusted weighted frequency of other types of characters in historical data;/> Indicates the initial frequency of characters of other character types in historical data;/> represents the normalization coefficient.

At this point, the adjusted weighted frequency of each type of character in historical data is obtained.

According to the adjusted weighted frequency of each type of character in the historical data, the frequency intervals of all types of characters in the historical data are arranged in descending order according to the adjusted weighted frequency. If there are weighted frequencies of the same size, the frequency intervals are arranged according to the ASCII value of the dictionary. Sort; then obtain the frequency intervals corresponding to all types of characters in the historical data, and use arithmetic coding to compress the newly added data.

The above descriptions are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the present invention. within the scope of protection.

Claims (10)

1. An artificial intelligence based intelligent medical data analysis method is characterized by comprising the following steps:

acquiring historical data, newly-added data type characters and newly-added data;

obtaining an initial frequency contribution weight of each character position in the historical data according to the data length of the historical data;

acquiring the occurrence frequency of newly added data in historical data; obtaining a final frequency contribution weight of each character position in the history data according to the character position of the newly added data in the history data, the initial frequency contribution weight of the character position and the frequency of the newly added data in the history data; obtaining the weighted frequency of the newly added data type character of the historical data according to the final frequency contribution weight of each character position in the historical data and the newly added data type character; obtaining an adjustment factor of the new data type character of the history data according to the number of the new data in the history data; obtaining final weighting frequency of the new data character type of the historical data according to the weighting frequency of the new data type character of the historical data and the adjustment factor of the new data type character of the historical data;

obtaining a normalization coefficient according to the final weighting frequency of the character type of the newly added data of the historical data; obtaining the weighted frequency of other types of characters of the historical data after adjustment according to the normalization coefficient; obtaining the weighted frequency after the adjustment of all types of characters of the historical data according to the final weighted frequency of the character types of the newly added data of the historical data and the weighted frequency after the adjustment of other types of characters of the historical data; and compressing the newly added data according to the weighted frequency of all types of characters of the historical data.

2. The intelligent medical data analysis method based on artificial intelligence according to claim 1, wherein the steps of obtaining history data, newly-added data type characters and newly-added data include the following steps:

acquiring historical intelligent medical data through medical equipment; after the historical intelligent medical data is obtained, adding new intelligent medical data at different times, and storing the new intelligent medical data and the historical intelligent medical data at the same time to obtain the new intelligent medical data; the collected historical intelligent medical data is recorded as historical data, character types of newly-added intelligent medical data are obtained and recorded as newly-added data type characters, and the newly-added intelligent medical data are recorded as newly-added data.

3. The intelligent medical data analysis method based on artificial intelligence according to claim 1, wherein the step of obtaining the initial frequency contribution weight of each character position in the history data according to the data length of the history data comprises the following specific steps:

the calculation expression for obtaining the initial frequency contribution weight of each character position in the historical data according to the poisson equation and the data length of the historical data is as follows:

in the method, in the process of the invention,representing->Initial frequency contribution weights for the individual character positions; />A data length representing the history data; lambda is the poisson parameter; />() An exponential function based on a natural constant is represented.

4. The intelligent medical data analysis method based on artificial intelligence according to claim 1, wherein the obtaining the final frequency contribution weight of each character position in the history data according to the character position of the newly added data in the history data, the initial frequency contribution weight of the character position and the frequency of the newly added data in the history data comprises the following specific steps:

for the ith character position in the historical data, according to the zero-variable of the ith character position and the initial frequency contribution weight, the calculation expression of the final frequency contribution weight of the ith character position in the historical data is as follows:

in the method, in the process of the invention,representing->Final frequency contribution weights for the individual character positions; />Representing->Zero-variable of the individual character positions; />Representing->Initial frequency contribution weights for the individual character positions; />A data length representing the history data;

the method for acquiring the zero-variable of the ith character position comprises the following steps: for any character of the newly added data, the character of the newly added data is corresponding to the ith character position in the historical data, and if the character on the ith character position is consistent with the character of the newly added data, the zero-variable V of the ith character position in the historical data is assigned as 1; if the character at the ith character position is inconsistent with the character of the newly added data, the zero-variable V at the ith character position in the history data is assigned as 0.

5. The intelligent medical data analysis method based on artificial intelligence according to claim 1, wherein the step of obtaining the weighted frequency of the new data type character of the history data according to the final frequency contribution weight of each character position in the history data and the new data type character comprises the following specific steps:

for the newly added data character type, according to the frequency of occurrence of the type character in the historical data, the character position and the final frequency contribution weight of the character position, the calculation expression of the weighting frequency of the type character in the historical data is as follows:

in the method, in the process of the invention,a weighted frequency representing the type of the newly added data character of the history data; />Representing->Final frequency contribution weights for the individual character positions; />Representing the frequency of the type of the newly added data character of the history data in the history data;the number of the character types of the newly added data in the history data is represented; />Representing the data length of the history data.

6. The intelligent medical data analysis method based on artificial intelligence according to claim 1, wherein the step of obtaining the adjustment factor of the new data type character of the history data according to the number of the new data appearing in the history data comprises the following specific steps:

for the newly added data type character, according to the number of the type character appearing in the historical data, acquiring a calculation expression of an adjustment factor of the type character as follows:

in the method, in the process of the invention,an adjustment factor representing a type of the newly added data character of the history data; />The number of the character types of the newly added data in the history data is represented; />A data length representing the history data; />A number of character types representing the history data; />() An exponential function based on a natural constant is represented.

7. The intelligent medical data analysis method based on artificial intelligence according to claim 1, wherein the obtaining the final weighted frequency of the added data character type of the history data according to the weighted frequency of the added data type character of the history data and the adjustment factor of the added data type character of the history data comprises the following specific steps:

taking the product of the adjustment factor of the added data character type of the historical data and the weighting frequency of the added data character type of the historical data as the final weighting frequency of the added data character type of the historical data.

8. The intelligent medical data analysis method based on artificial intelligence according to claim 1, wherein the specific formula for obtaining the normalized coefficient according to the final weighted frequency of the added data character type of the history data is as follows:

in the method, in the process of the invention,representing the normalization coefficient; />A final weighting frequency representing the newly added data character type of the history data; />The frequency of the type of the newly added data character in the history data is represented.

9. The intelligent medical data analysis method based on artificial intelligence according to claim 1, wherein the step of obtaining the weighted frequency of the other types of characters of the history data after adjustment according to the normalization coefficient comprises the following specific steps:

taking the product of the initial frequency of the characters of other character types in the historical data and the normalized coefficient as the weighted frequency of the characters of other types in the historical data after adjustment.

10. The intelligent medical data analysis method based on artificial intelligence according to claim 1, wherein the compressing the newly added data according to the weighted frequency of all types of characters of the history data comprises the following specific steps:

according to the weighted frequency of each type of character of the historical data after adjustment, the frequency intervals of all types of characters of the historical data are arranged according to the order from big to small of the weighted frequency after adjustment, and if the weighted frequencies with the same size exist, the frequency intervals are arranged according to the ASCII values of the dictionary; obtaining frequency intervals corresponding to all types of characters of the historical data, and compressing the newly added data by utilizing arithmetic coding.