CN115640585A - Data processing method and device and readable storage medium - Google Patents
Data processing method and device and readable storage medium Download PDFInfo
- Publication number
- CN115640585A CN115640585A CN202110817297.0A CN202110817297A CN115640585A CN 115640585 A CN115640585 A CN 115640585A CN 202110817297 A CN202110817297 A CN 202110817297A CN 115640585 A CN115640585 A CN 115640585A
- Authority
- CN
- China
- Prior art keywords
- data
- ciphertext
- encrypted
- generation model
- language generation
- 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
Images
Landscapes
- Storage Device Security (AREA)
Abstract
The invention provides a data processing method, which comprises the following steps: acquiring data to be queried; converting the data to be inquired into a coded value of a preset coding format; encrypting the coded value by adopting a first encryption algorithm to obtain first encrypted data, inputting the first encrypted data into a language generation model, and obtaining a statement set generated by the language generation model aiming at statement association of the first encrypted data; encrypting each element in the sentence set by adopting a second encryption algorithm to obtain a plurality of data ciphertexts to be inquired; and matching the plurality of data ciphertexts to be inquired with the cipher text database one by one, and decrypting the target cipher text into a plaintext to be returned as an inquiry result under the condition that the target cipher text corresponding to the data ciphertexts to be inquired exists in the cipher text database.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a data processing method and apparatus, and a readable storage medium.
Background
In order to improve the security of data in the transmission process, data can be encrypted first, and at present, when the data is encrypted, a word segmentation combination is performed on the data needing to be encrypted, and result sets of the word segmentation combination are encrypted respectively and then stored in an expansion column. For example: characters are grouped in fixed length, and one field is divided into multiple groups. When the method is adopted for encryption, the fixed length of the characters must be specified, and the length of the query characters also needs to be specified during query, so that the ciphertext is easy to crack, and the encryption safety is low.
Disclosure of Invention
The embodiment of the invention provides a data processing method, a data processing device and a readable storage medium, which aim to solve the problem of low encryption security in the prior art.
In order to solve the technical problem, the invention is realized as follows:
in a first aspect, an embodiment of the present invention provides a data processing method, including:
acquiring data to be queried;
converting the data to be inquired into a coded value of a preset coding format;
encrypting the coded value by adopting a first encryption algorithm to obtain first encrypted data, inputting the first encrypted data into a language generation model, and obtaining a statement set generated by the language generation model by statement association aiming at the first encrypted data;
encrypting each element in the statement set by adopting a second encryption algorithm to obtain a plurality of data ciphertexts to be inquired;
and matching the plurality of data ciphertexts to be inquired with a cipher text database one by one, and decrypting the target cipher text into a plaintext as an inquiry result to return under the condition that the target cipher text corresponding to the data ciphertexts to be inquired exists in the cipher text database.
In a second aspect, an embodiment of the present invention provides a data processing apparatus, including:
the acquisition module is used for acquiring data to be queried;
the conversion module is used for converting the data to be inquired into a code value in a preset code format;
the first encryption module is used for encrypting the coded value by adopting a first encryption algorithm to obtain first encrypted data, inputting the first encrypted data into a language generation model, and obtaining a statement set generated by the language generation model aiming at statement association of the first encrypted data;
the second encryption module is used for encrypting each element in the statement set by adopting a second encryption algorithm to obtain a plurality of data ciphertexts to be inquired;
and the query module is used for matching the plurality of data ciphertexts to be queried with the ciphertext database one by one, and decrypting the target ciphertext into a plaintext to be returned as a query result under the condition that the target ciphertext corresponding to the data ciphertexts to be queried exists in the ciphertext database.
In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, where a computer program is stored, and when being executed by a processor, the computer program implements the steps of the data processing method according to the first aspect.
According to the data processing method, during the data query process, the ciphertext form is used for query, the safety in the data query process is guaranteed, sentence association is conducted on data to be queried based on the language generation model, further the sentence set generated by the sentence association is matched with the ciphertext database one by one, accurate search and matching can be achieved, and the accuracy of query results is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive exercise.
Fig. 1 is a flowchart of a data processing method according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating a case of a word segmentation group according to an embodiment of the present invention;
fig. 3 is a second flowchart of a data processing method according to an embodiment of the present invention;
fig. 4 is a block diagram of a data processing apparatus according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without any inventive step, are within the scope of the present invention.
Referring to fig. 1, fig. 1 is a data processing method provided by an embodiment of the present invention, including:
In this step, the data to be queried refers to data that the user needs to query, and this data may be a keyword or a phrase, which is only an example and is not limited herein.
In this embodiment, the data to be queried may be denoted as "j1j2.. Jk".
And 102, converting the data to be inquired into a code value in a preset code format.
Specifically, the preset encoding format may be an American Standard Code for Information Interchange (ASCII) encoding format, and when the data to be encrypted is converted into the ASCII encoding format, each character in the data to be encrypted is converted into an ASCII value in the ASCII encoding format, where one chinese character may represent one character, one english character may represent one character, and one numeral may also be one character.
103, encrypting the coded value by adopting a first encryption algorithm to obtain first encrypted data, inputting the first encrypted data into a language generation model, and obtaining a statement set generated by the language generation model by statement association aiming at the first encrypted data.
In this step, the first encryption algorithm may be an order-preserving encryption algorithm (which may be abbreviated as BCLO-09 algorithm), which is only an example and is not limited herein. Alternatively, in other possible embodiments, other types of encryption algorithms may be used, and no matter how the encryption algorithm is changed, the protection scope of the embodiments of the present application is within the scope of the present application.
Specifically, in this step, the language generation model may be an N-gram algorithm model, which is only an example and is not limited herein, and the language generation model may be used for performing character association, for example, a sentence that i need to query is "12345", and is "abcde" after encryption, and only by inputting "ab" into the language generation model, the language generation model can automatically generate "abcde", and then an effect of accurate matching can be achieved from fuzzy matching, so that the purpose of querying "abcd" through "ab" is achieved.
And step 104, encrypting each element in the statement set by adopting a second encryption algorithm to obtain a plurality of data ciphertexts to be inquired.
In this step, the second encryption algorithm may be a reversible encryption algorithm, which is only an example and is not limited herein. Alternatively, in other possible embodiments, other types of encryption algorithms may be used, but any transformation thereof is within the scope of the embodiments of the present application.
And 105, matching the plurality of data ciphertexts to be queried with a ciphertext database one by one, and decrypting a target ciphertext into a plaintext as a query result to return under the condition that the target ciphertext corresponding to the data ciphertext to be queried exists in the ciphertext database.
And accurately matching the plurality of data ciphertexts to be inquired obtained by the second encryption with the ciphertexts in the cipher text database, and if the matching is successful, decrypting the successfully matched ciphertexts, wherein the decryption process can comprise the following steps.
Firstly, the ciphertext successfully matched is decrypted for the first time by adopting a reversible encryption algorithm, then the decrypted result of the first time is decrypted for the second time by an order-preserving encryption algorithm, and then the decrypted result of the second time is subjected to ASCII transcoding to obtain the final plaintext 'J1J2.. Jk... Jn and Z1Z2.. Zn', and the plaintext is returned to the user.
According to the data processing method, during the data query process, the ciphertext form is used for querying, the safety during the data query process is guaranteed, sentence association is carried out on the data to be queried based on the language generation model, further, the sentence set generated by the sentence association is matched with the ciphertext database one by one, the accurate search and matching can be realized, and the accuracy of the query result is improved.
Optionally, the data processing method further includes:
storing data to be queried, a plaintext and a corresponding relation between the data to be queried and the plaintext into a cache region;
converting the data to be queried into an encoding value of a preset encoding format, comprising:
and matching the data to be queried with the stored plain text in the cache region, and converting the data to be queried into a code value in a preset code format under the condition of unsuccessful matching.
In this optional embodiment, since the user may need to search the ciphertext database for multiple times, in order to improve the search speed in the subsequent search process, when querying the data to be queried, if the query is successful in matching the corresponding ciphertext, after converting the ciphertext into the plaintext and returning the plaintext to the user, the data to be queried, the plaintext, and the correspondence between the data to be queried and the plaintext are stored in the cache region.
Optionally, in a case that a target ciphertext corresponding to the data ciphertext to be queried does not exist in the ciphertext database, the data processing method further includes:
acquiring characters which fail to be matched;
and reducing the parameter weight of the character which is matched in failure in the language generation model according to the character which is matched in failure.
In this optional embodiment, if there is no target ciphertext corresponding to the ciphertext of the data to be queried in the ciphertext database, the characters included in the statement output by the language generation model during the encryption process of the data to be queried are obtained, and the parameter weight of the characters corresponding to the output result of the language output model is reduced according to the characters, so as to further optimize the language generation model.
Optionally, before the ciphertext of the data to be queried is matched with the ciphertext database one by one, the data processing method further includes:
acquiring data to be encrypted;
and constructing a ciphertext database according to the data to be encrypted.
Optionally, the constructing a ciphertext database according to the data to be encrypted includes:
converting data to be encrypted into a coded value in a preset coding format;
encrypting the coded value by adopting a first encryption algorithm to obtain second encrypted data, and performing word segmentation combination on the second encrypted data to generate a first data set;
and encrypting the first data set by adopting a second encryption algorithm to obtain a ciphertext, and constructing a ciphertext database according to the ciphertext.
In this optional embodiment, during encryption, data to be encrypted is encrypted, and a ciphertext database is constructed according to the encrypted data, so that in subsequent search, search can be performed from the ciphertext database.
Optionally, before constructing the ciphertext database according to the ciphertext, the data processing method further includes:
allocating a unique corresponding index for the ciphertext;
and constructing a ciphertext database according to the ciphertext and the index.
It should be appreciated that the index may be a tag that uniquely corresponds to the ciphertext by which the corresponding ciphertext may be quickly queried. In the optional embodiment, the only corresponding index is allocated to the ciphertext, and the ciphertext and the index are stored in the ciphertext database in an associated manner, so that the ciphertext can be conveniently and quickly queried according to the index when the ciphertext is queried subsequently, and the query efficiency is improved.
Optionally, the method further includes:
and training to obtain a language generation model based on the first data set and the second encrypted data.
Optionally, training a language generation model based on the first data set and the second encrypted data includes:
inputting a first data set into an initial language generation model, and acquiring a second data set output by the initial language generation model, wherein the first data set comprises target elements, and the second data set comprises target statements obtained based on the target elements;
matching the target statement with the encrypted data;
under the condition that the second data set is not matched with the encrypted data, adjusting the parameter weight in the initial language generation model to perform iterative training on the initial language generation model so as to obtain a language generation model when the iterative training is finished;
and the adjusted parameter weight is used for enabling the initial language generation model not to contain the target sentence in the output result of the target element in the next iteration training.
In this alternative embodiment, the second data set is the output of the language generation model when the initial language generation model is trained with the first data set and the second encrypted data to obtain the final language generation model. It should be understood that the number of the target elements may be 1 or more, for example, when the first data set is P, the target element may be P1, and the second data set is a target sentence generated by the language generation model according to the target element and including the characters Y1, Y2 … Yn.
It should be understood that when the encrypted data is subjected to word segmentation, if only one character exists in the encrypted data, the word segmentation does not need to be carried out. If the number of characters in the encrypted data is greater than or equal to 2, two characters may be used as a group to perform word segmentation and combination on the encrypted data, for example, the code value of the data to be encrypted is "a1a2a2a3.. Am", the encrypted data obtained after being encrypted by the order-preserving encryption algorithm is "b1b2b3.. Bm", "c1c2c3.. Cm", where m represents the total number of data, "b1b2b3.. Bm" corresponds to one encrypted character, "c1c2c3.. Cm" corresponds to one encrypted character, and the result of performing word segmentation and combination on the second encrypted data may be "B1B2", "B2B3", "B (m-1) Bm", "C1C2", "C2C3", "C (m-1) Cm", and the like, which is only used for example and is not limited herein. In other possible embodiments, other types of word segmentation combinations may also be performed on the second encrypted data, for example, the result of the word segmentation combination may be "B1B2B3", "B (m-1) Bm B (m + 1)", "C1C2C3", and the like. The schematic diagram of the word segmentation combination situation is shown in fig. 2, and further, a first data set P is generated according to the result of the word segmentation combination, where the first data set P may be represented as "{ (B1B 2), (B2B 3) … (B (m-1) Bm) }" or "{ (C1C 2), (C2C 3) … (C (m-1) Cm) }". Further, all the encrypted data described above may be represented by the second encrypted data T.
Further, Y1, Y2 … Yn and the second encrypted data T are accurately matched, if the character Y1 cannot be matched with the element in the second encrypted data T, the parameter weight corresponding to the Y1 in the language generation model is reduced, and therefore when the character Y1 is input into the language generation model again next time, the sentence output by the language generation model does not include the character Y1 any more, but only includes the character Y2 … Yn. In this way, by matching the elements in the generated sentence with the elements in the encrypted data set, the parameter weight corresponding to the generated elements in the language generation model is adjusted, and the output of the language generation model can be more accurate.
It should be noted that, in the present embodiment, a ciphertext query mode is adopted, so that the security of data in the query process can be ensured, and therefore, the data to be queried is encrypted by the above method to obtain a ciphertext of the data to be queried. For example, the data "j1j2.. Jk" is encrypted by an order-preserving algorithm to obtain "e1e2.. Ek", and then a statement D corresponding to "e1e2.. Ek" is generated by a language generation model, and the statement D includes "{ e1e2.. En, q1q2.. Qn, and nj1n2.. Nn.. }", where "q1q2.. Qn" and "nj1n2.. Nn" may represent statements having similar meanings obtained from input characters. Further, all elements in statement D may be encrypted a second time by a reversible encryption algorithm.
It should be noted that, in order to ensure the accuracy of the language generation model, in this optional embodiment, the first data set and the second encrypted data at each encryption may be used as training data for training the language generation model, so that the training sample of the language generation model may be expanded, in other words, the new first data set represents a data set formed by combining the encryption results of the data to be encrypted, which needs to be encrypted when the data encryption method is used for encryption next time. In this way, by enlarging the training sample of the language generation model, the accuracy of the language generated by the trained language generation model can be improved. It is worth emphasizing that, in the optional embodiment, in the process of training the language generation model, a value of N in the N-gram algorithm model may be adjusted, where N represents the number of each group of participles in the first data set, so that effective characters in the first data set may be accurately learned, and a complete sentence may be quickly output according to the first data set without limiting the length of the characters. In another embodiment, as shown in fig. 3, a method for processing data is described as an example, first, data to be encrypted is acquired, and the data to be encrypted is encrypted, wherein the encrypting step includes converting the data into ASCII values, then encrypting the ASCII values for the first time by using a order-preserving encryption algorithm, generating a sentence from the first encryption result by using a pre-trained language generation model, encrypting the generated sentence for the second time by using a reversible encryption algorithm, and then adding an index to the encryption result and storing the encryption result in a ciphertext database.
Further, a data query method is executed, wherein the data query step comprises querying a plaintext corresponding to data to be queried in a cache region, if the query is unsuccessful, converting the data to be queried into an ASCII value, then performing first encryption on the ASCII value by using an order-preserving encryption algorithm, generating a statement according to a first encryption result by using a pre-trained language generation model, performing second encryption on the generated statement by using a reversible encryption algorithm, accurately matching the second encryption result with the ciphertext database, if the matching fails, adjusting the language generation model, if the target ciphertext is successfully matched, performing first decryption on the target ciphertext by using a reversible encryption algorithm, then performing second decryption by using the order-preserving encryption algorithm, then performing ASCII transcoding to obtain the plaintext, and storing the plaintext, the data to be queried and a corresponding relationship between the plaintext and the data to be queried in the cache region into the cache region. Therefore, the process of data encryption and data query once is realized, the data is encrypted through the encryption process twice, the data security can be improved, and the data security in the query process can be ensured by adopting ciphertext query during query.
Referring to fig. 4, an embodiment of the present application further provides a data processing apparatus, including:
the acquisition module is used for acquiring data to be inquired;
the conversion module is used for converting the data to be inquired into a code value in a preset code format;
the first encryption module is used for encrypting the coding value by adopting a first encryption algorithm to obtain first encrypted data, inputting the first encrypted data into a language generation model, and acquiring a statement set generated by the language generation model aiming at statement association of the first encrypted data;
the second encryption module is used for encrypting each element in the statement set by adopting a second encryption algorithm to obtain a plurality of data ciphertexts to be inquired;
and the query module is used for matching the plurality of data ciphertexts to be queried with the ciphertext database one by one, and decrypting the target ciphertext into a plaintext to be returned as a query result under the condition that the target ciphertext corresponding to the data ciphertexts to be queried exists in the ciphertext database.
The data processing apparatus can implement the embodiments of the data processing method and achieve the same beneficial effects, and details are not repeated here.
The embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements each process of the above-mentioned network slice creation method embodiment, and can achieve the same technical effect, and in order to avoid repetition, the computer program is not described here again. The computer-readable storage medium may be a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.
Through the above description of the embodiments, those skilled in the art will clearly understand that the above embodiment method can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
While the present invention has been described with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, which are illustrative and not restrictive, and it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (10)
1. A data processing method, comprising:
acquiring data to be queried;
converting the data to be inquired into a coded value of a preset coding format;
encrypting the coded value by adopting a first encryption algorithm to obtain first encrypted data, inputting the first encrypted data into a language generation model, and obtaining a statement set generated by the language generation model by statement association aiming at the first encrypted data;
encrypting each element in the statement set by adopting a second encryption algorithm to obtain a plurality of data ciphertexts to be inquired;
and matching the plurality of data ciphertexts to be inquired with a cipher text database one by one, and decrypting the target cipher text into a plain text as an inquiry result to return under the condition that the target cipher text corresponding to the data ciphertexts to be inquired exists in the cipher text database.
2. The data processing method of claim 1, wherein the method further comprises:
storing the data to be queried, the plaintext and the corresponding relationship between the data to be queried and the plaintext into a cache region;
the converting the data to be queried into an encoding value of a preset encoding format includes:
and matching the data to be queried with the plaintext stored in the cache region, and converting the data to be queried into a code value in a preset code format under the condition of unsuccessful matching.
3. The data processing method according to claim 1, wherein in a case where a target ciphertext corresponding to the data ciphertext to be queried does not exist in the ciphertext database, the method further comprises:
acquiring characters which fail to be matched;
and reducing the parameter weight of the character which is in the language generation model and corresponds to the matching failure according to the character which is in the matching failure.
4. The data processing method according to claim 1, wherein before matching the plurality of data ciphertexts to be queried with the ciphertext database one by one, the method further comprises:
acquiring data to be encrypted;
and constructing a ciphertext database according to the data to be encrypted.
5. The data processing method according to claim 4, wherein the constructing a ciphertext database according to the data to be encrypted comprises:
converting the data to be encrypted into an encoded value of the preset encoding format;
encrypting the coded value by adopting the first encryption algorithm to obtain second encrypted data, and performing word segmentation combination on the second encrypted data to generate a first data set;
and encrypting the first data set by adopting the second encryption algorithm to obtain a ciphertext, and constructing a ciphertext database according to the ciphertext.
6. The data processing method of claim 5, wherein before the constructing the ciphertext database from the ciphertext, the method further comprises:
allocating a unique corresponding index to the ciphertext;
and constructing the ciphertext database according to the ciphertext and the index.
7. The data processing method of claim 5, wherein the method further comprises:
and training to obtain the language generation model based on the first data set and the second encrypted data.
8. The data processing method of claim 7, wherein training the language generation model based on the first data set and the second encrypted data comprises:
inputting the first data set into an initial language generation model, and acquiring a second data set output by the initial language generation model, wherein the first data set comprises target elements, and the second data set comprises target statements obtained based on the target elements;
matching the target statement with the encrypted data;
under the condition that the second data set is not matched with the encrypted data, adjusting parameter weights in the initial language generation model to perform iterative training on the initial language generation model so as to obtain the language generation model when the iterative training is finished;
wherein the adjusted parameter weight is used for enabling the initial language generation model not to contain the target statement in an output result of the target element in next iteration training.
9. A data processing apparatus, comprising:
the acquisition module is used for acquiring data to be inquired;
the conversion module is used for converting the data to be inquired into a code value in a preset code format;
the first encryption module is used for encrypting the coded value by adopting a first encryption algorithm to obtain first encrypted data, inputting the first encrypted data into a language generation model, and obtaining a statement set generated by the language generation model by statement association aiming at the first encrypted data;
the second encryption module is used for encrypting each element in the statement set by adopting a second encryption algorithm to obtain a plurality of data ciphertexts to be inquired;
and the query module is used for matching the plurality of data ciphertexts to be queried with the ciphertext database one by one, and decrypting the target ciphertext into a plaintext to be returned as a query result under the condition that the target ciphertext corresponding to the data ciphertexts to be queried exists in the ciphertext database.
10. A readable storage medium, characterized in that it stores thereon a program or instructions which, when executed by a processor, implement the steps of the data processing method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110817297.0A CN115640585A (en) | 2021-07-20 | 2021-07-20 | Data processing method and device and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110817297.0A CN115640585A (en) | 2021-07-20 | 2021-07-20 | Data processing method and device and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115640585A true CN115640585A (en) | 2023-01-24 |
Family
ID=84940389
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110817297.0A Pending CN115640585A (en) | 2021-07-20 | 2021-07-20 | Data processing method and device and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115640585A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117331970A (en) * | 2023-10-31 | 2024-01-02 | 中科驭数（北京）科技有限公司 | Data query method, device, computer storage medium and acceleration card |
-
2021
- 2021-07-20 CN CN202110817297.0A patent/CN115640585A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117331970A (en) * | 2023-10-31 | 2024-01-02 | 中科驭数（北京）科技有限公司 | Data query method, device, computer storage medium and acceleration card |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200342183A1 (en) | Processing text sequences using neural networks | |
| US10515155B2 (en) | Conversational agent | |
| CN108334612B (en) | Shape-near Chinese character full-text fuzzy retrieval method aiming at ciphertext domain | |
| CN111814466A (en) | Information extraction method based on machine reading understanding and related equipment thereof | |
| CN109726585B (en) | Integrated data desensitization system and method for identity card number | |
| CN106610995B (en) | Method, device and system for creating ciphertext index | |
| CN111797409B (en) | Carrier-free information hiding method for big data Chinese text | |
| Moraldo | An approach for text steganography based on markov chains | |
| CN112257455B (en) | Semantic understanding ciphertext space keyword retrieval method and system | |
| CN110222008A (en) | A kind of conversion method of data format, system, medium and electronic equipment | |
| CN115880826B (en) | Park access method and system based on access data | |
| CN111651986A (en) | Event keyword extraction method, device, equipment and medium | |
| CN113255007A (en) | Safe hidden three-factor query method | |
| CN113656547A (en) | Text matching method, device, equipment and storage medium | |
| EP3525107A1 (en) | Conversational agent | |
| CN113408278A (en) | Intention recognition method, device, equipment and storage medium | |
| CN115640585A (en) | Data processing method and device and readable storage medium | |
| CN115908641A (en) | Text-to-image generation method, device and medium based on features | |
| CN111709229A (en) | Text generation method and device based on artificial intelligence, computer equipment and medium | |
| KR20220152167A (en) | A system and method for detecting phishing-domains in a set of domain name system(dns) records | |
| Safaka et al. | Matryoshka: Hiding secret communication in plain sight | |
| CN116825259B (en) | Medical data management method based on Internet of things | |
| CN112966086A (en) | Verifiable fuzzy search method based on position sensitive hash function | |
| CN112711648A (en) | Database character string ciphertext storage method, electronic device and medium | |
| CN114297693B (en) | Model pre-training method and device, electronic equipment and storage medium |
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 |