CN106899572A - Sterility testing data staging encryption method based on condition random field algorithm - Google Patents

Abstract

The invention discloses a classification encryption method for aseptic detection data based on a conditional random field algorithm. Firstly, the important result data generated in the aseptic detection process is marked, and the conditional random field model is trained to identify the important result data; secondly, the important result data is identified. The important result data identified by the conditional random field model is encrypted hierarchically, and the encrypted data is transmitted to the receiver; finally, the receiver uses a reversible decryption process to restore the data without loss. The invention uses a conditional random field method to identify important result data generated in the aseptic testing process, and distinguishes it from process data for hierarchical encryption. At the same time, digital signature technology is used to effectively prevent messages from being tampered with, ensuring that data will not be leaked. On the premise of ensuring the security of important data information, the invention can greatly reduce the encryption time and improve the system efficiency.

Description

A Hierarchical Encryption Method for Aseptic Testing Data Based on Conditional Random Field Algorithm

technical field

The invention belongs to the field of microbial sterility detection, and in particular relates to a method for hierarchical encryption of sterility detection data based on a conditional random field algorithm.

Background technique

In industrial production, microbial sterility testing is of great significance to ensure product safety. In recent years, with the development of informatization and automation industries, domestic research in the field of sterility testing has gradually tended to be automated. However, the level of intelligence in microbial sterility testing is low, the use of information technology is insufficient, and data integrity, security, and traceability are poor. In order to ensure the security of the data generated in the process of sterility testing, it is necessary to encrypt the important data in order to prevent the leakage or artificial tampering of data information. However, not all data is meaningful in the entire sterility testing system. If a large amount of process data is encrypted together with important result data, it will take a lot of time and reduce system efficiency. The existing data encryption method mixes all process data and result data together for encryption, but in practical applications, only the result data is often used. A large amount of process data will increase the time of encryption and decryption, which will affect the efficiency of the system. With the increase of the data volume of the aseptic testing platform, more and more time will be spent due to redundant encryption, and the efficiency of the system will be seriously affected. .

Contents of the invention

The object of the present invention is to aim at the deficiencies in the prior art, provide a kind of sterility detection data graded encryption method based on the conditional random field algorithm, the important result that will produce in the automatic sterility detection process by the method of conditional random field (CRF) The data is identified, and the resulting data and process data are encrypted hierarchically using symmetric encryption technology, asymmetric encryption technology, and digital summary technology, which essentially solves the drawbacks of consuming a lot of time due to too much encrypted data. Compared with existing methods, It has the characteristics of safety and high efficiency. Complex encryption of the most important data and relatively simple encryption of other data will also make the system more flexible, minimizing costs and saving time.

The object of the present invention is achieved by the following technical scheme: a kind of sterility detection data classification encryption method based on conditional random field algorithm, comprises the following steps:

(1) Mark the important result data generated in the sterility testing process, train the conditional random field model, and identify the important result data;

(2) Perform high-level RSA encryption on the important result data identified by the conditional random field model, perform low-level DES encryption on ordinary process data, and transmit the encrypted data to the receiver;

(3) The receiver uses a reversible decryption process to restore the data losslessly.

Further, the step 1 specifically includes the following sub-steps:

(1.1) select the data that an amount of aseptic testing process produces, equally divide into two groups as training corpus and test corpus respectively;

(1.2) Carry out word segmentation processing with two groups of corpora;

(1.3) Carry out BIO labeling to the corpus processed through word segmentation, mark out important result data as named entities, and obtain the labeled data set;

(1.4) feature extraction, select single character, word, part-of-speech tagging and correct spelling as feature set, obtain conditional random field model with training corpus training;

(1.5) Test and correct the model obtained in step 1.4 with the test corpus to obtain a conditional random field model that can accurately identify important result data;

(1.6) For the data generated in the sterility testing process, important result data are identified through the model obtained in step 1.5.

Further, the step 2 specifically includes the following sub-steps:

(2.1) Divide the identified data into two categories: important result data and common process data;

(2.2) Hash the result data to obtain the digital summary MD, and perform the Hash operation on the process data to obtain the digital summary MD';

(2.3) Select n in the set (2 ²⁰⁴⁷ , 2 ²⁰⁴⁸ ], satisfying n=p×q, wherein p and q are both prime numbers;

(2.4) Choose e to make it the same as is relatively prime and less than

(2.5) by formula Choose a large integer less than 2048bits;

(2.6) Use {e, n} as the public key, use the RSA algorithm to encrypt the digital abstract MD, MD', and obtain the digital signature DS, DS';

(2.7) Use the public key of RSA to encrypt the result data and digital signature DS to obtain encrypted information E, and use the DES algorithm to encrypt the process data and digital signature DS' to obtain encrypted information E';

(2.8) Pack the encrypted result data and process data into digital envelopes and send them to the receiver.

Further, the step 3 specifically includes the following sub-steps:

(3.1) Receive digital envelopes;

(3.2) Use {d,n} as the private key to decrypt E with RSA to obtain the result data and digital signature DS, and use the DES algorithm to decrypt E' to obtain process data and digital signature DS';

(3.3) Use the same Hash operation on the result data to obtain the digital summary md, and use the same Hash operation on the process data to obtain the digital summary md';

(3.4) Use {d, n} as the private key to decrypt DS with RSA to obtain MD, and use {d, n} as the private key to decrypt DS’ with RSA to obtain MD’;

(3.5) Compare the two digital summaries MD and md, MD' and md', and verify whether the original data has been modified. If the two are the same, it means that the data has not been tampered with, is transmitted confidentially, and the signature is authentic; otherwise, the signature is rejected.

The beneficial effects of the present invention are: the present invention uses the conditional random field method to identify important result data generated in the aseptic testing process, and distinguishes it from process data for hierarchical encryption. The relatively complex RSA encryption is used for the result data to improve data security; the relatively simple DES encryption is used for the process data to improve efficiency; at the same time, the digital signature technology is used to effectively prevent the message from being tampered with and ensure that the data will not be leaked. On the premise of ensuring the security of important data information, the invention can greatly reduce the encryption time and improve the system efficiency.

Description of drawings

Figure 1 is a flow chart of identifying important result data by constructing a conditional random field model;

Fig. 2 is a flowchart of the encryption process;

Figure 3 is a flowchart of the decryption process.

detailed description

The present invention will be described in further detail below in conjunction with the accompanying drawings and specific embodiments.

A kind of aseptic detection data hierarchical encryption method based on conditional random field algorithm provided by the present invention comprises the following steps:

(1) Build a conditional random field model to identify important result data, as shown in Figure 1, specifically including the following sub-steps:

(1.1) select the data that an amount of aseptic testing process produces, equally divide into two groups as training corpus and test corpus respectively;

(1.2) Segment the two groups of corpora; word segmentation refers to dividing a sequence of Chinese characters into individual words, which is the process of recombining continuous word sequences into word sequences according to certain norms. We know that in English writing, spaces are used as natural delimiters between words, but in Chinese, only words, sentences and paragraphs can be delimited by obvious delimiters, except that words do not have a formal delimiter , so Chinese word segmentation is very basic and critical in the processing of Chinese information;

(1.3) Carry out BIO annotation on the corpus processed by word segmentation, annotate important result data as named entities, and obtain annotated data set; here, "B" represents the beginning of an entity, "I" represents the content of the entity, and " O" stands for Outer Entity. An example of an annotation entity annotated with BIO is shown in the following table:

(1.4) Feature extraction, select single characters, words, part-of-speech tagging and correct spelling as feature sets, and use the training corpus to train the conditional random field model; where single characters mean single Chinese characters, punctuation marks, foreign letters and numbers. The main characteristic of Chinese narrative text is that it is composed of a series of consecutive Chinese characters without spaces to separate them. Therefore, we also choose words as another feature. Selecting part-of-speech tagging as the feature of machine learning has a good effect on the recognition of the boundaries of Chinese entities. The correct spelling feature is used to identify entities composed of English.

(1.5) test and correct the model obtained in step 1.4) with the test corpus, and obtain a conditional random field model that can accurately identify important result data;

(1.6) For the data generated in the sterility testing process, important result data are identified through the model obtained in step 1.5).

(2) Encrypt the data generated in the sterility testing process, as shown in Figure 2, specifically comprising the following sub-steps:

(2.1) Divide the identified data into two categories: important result data and common process data;

(2.2) Hash the result data to obtain the digital summary MD, and perform the Hash operation on the process data to obtain the digital summary MD';

(2.3) Select n in the set (2 ²⁰⁴⁷ , 2 ²⁰⁴⁸ ], satisfying n=p×q, wherein p and q are both prime numbers;

(2.4) Choose e to make it the same as is relatively prime and less than Considering that it is relatively safe and fast, you can choose 65537 as the value of e;

(2.5) by formula Choose a large integer less than 2048bits;

(2.6) Use {e, n} as the public key, use the RSA algorithm to encrypt the digital abstract MD, MD', and obtain the digital signature DS, DS';

(2.7) Use the public key of RSA to encrypt the result data and digital signature DS to obtain the encrypted information E, and use the DES algorithm to encrypt the process data and digital signature DS' to obtain the encrypted information E';

(2.8) Pack the encrypted result data and process data into digital envelopes and send them to the receiver.

(3) The receiver uses the reversible decryption process to restore the data losslessly, as shown in Figure 3, which specifically includes the following sub-steps:

(3.1) Receive digital envelopes;

(3.2) Use {d,n} as the private key to decrypt E with RSA to obtain the result data and digital signature DS, and use the DES algorithm to decrypt E' to obtain process data and digital signature DS';

(3.3) Use the same Hash operation on the result data to obtain the digital summary md, and use the same Hash operation on the process data to obtain the digital summary md';

(3.4) Use {d, n} as the private key to decrypt DS with RSA to obtain MD, and use {d, n} as the private key to decrypt DS’ with RSA to obtain MD’;

(3.5) Compare the two digital summaries MD and md, MD' and md', and verify whether the original data has been modified. If the two are the same, it means that the data has not been tampered with, is transmitted confidentially, and the signature is authentic; otherwise, the signature is rejected.

Claims (4)

1. a kind of aseptic detection data hierarchical encryption method based on conditional random field algorithm, it is characterized in that, may further comprise the steps: (1) Mark the important result data generated in the sterility testing process, train the conditional random field model, and identify the important result data; (2) Perform high-level RSA encryption on the important result data identified by the conditional random field model, perform low-level DES encryption on ordinary process data, and transmit the encrypted data to the receiver; (3) The receiver uses a reversible decryption process to restore the data losslessly. 2. a kind of asepsis detection data hierarchical encryption method based on conditional random field algorithm according to claim 1, it is characterized in that, described step 1 specifically comprises the following sub-steps: (1.1) select the data that an amount of aseptic testing process produces, equally divide into two groups as training corpus and test corpus respectively; (1.2) Carry out word segmentation processing with two groups of corpora; (1.3) Carry out BIO labeling to the corpus processed through word segmentation, mark out important result data as named entities, and obtain the labeled data set; (1.4) feature extraction, select single character, word, part-of-speech tagging and correct spelling as feature set, obtain conditional random field model with training corpus training; (1.5) Test and correct the model obtained in step 1.4 with the test corpus to obtain a conditional random field model that can accurately identify important result data; (1.6) For the data generated in the sterility testing process, important result data are identified through the model obtained in step 1.5. 3. a kind of asepsis detection data hierarchical encryption method based on conditional random field algorithm according to claim 1, is characterized in that, described step 2 specifically comprises the following sub-steps: (2.1) Divide the identified data into two categories: important result data and common process data; (2.2) Hash the result data to obtain the digital summary MD, and perform the Hash operation on the process data to obtain the digital summary MD'; (2.3) Select n in the set (2 ²⁰⁴⁷ , 2 ²⁰⁴⁸ ], satisfying n=p×q, wherein p and q are both prime numbers; (2.4) Choose e to make it the same as is relatively prime and less than (2.5) by formula Choose a large integer less than 2048bits; (2.6) Use {e, n} as the public key, use the RSA algorithm to encrypt the digital abstract MD, MD', and obtain the digital signature DS, DS'; (2.7) Use the public key of RSA to encrypt the result data and digital signature DS to obtain encrypted information E, and use the DES algorithm to encrypt the process data and digital signature DS' to obtain encrypted information E'; (2.8) Pack the encrypted result data and process data into digital envelopes and send them to the receiver. 4. a kind of asepsis detection data hierarchical encryption method based on conditional random field algorithm according to claim 1, is characterized in that, described step 3 specifically comprises the following sub-steps: (3.1) Receive digital envelopes; (3.2) Use {d,n} as the private key to decrypt E with RSA to obtain the result data and digital signature DS, and use the DES algorithm to decrypt E' to obtain process data and digital signature DS'; (3.3) Use the same Hash operation on the result data to obtain the digital summary md, and use the same Hash operation on the process data to obtain the digital summary md'; (3.4) Use {d, n} as the private key to decrypt DS with RSA to obtain MD, and use {d, n} as the private key to decrypt DS’ with RSA to obtain MD’; (3.5) Compare the two digital summaries MD and md, MD' and md', and verify whether the original data has been modified. If the two are the same, it means that the data has not been tampered with, is transmitted confidentially, and the signature is authentic; otherwise, the signature is rejected.