CN110413958A - Linear congruence character set transform method and system for automatic machine space compression - Google Patents
Linear congruence character set transform method and system for automatic machine space compression Download PDFInfo
- Publication number
- CN110413958A CN110413958A CN201910505446.2A CN201910505446A CN110413958A CN 110413958 A CN110413958 A CN 110413958A CN 201910505446 A CN201910505446 A CN 201910505446A CN 110413958 A CN110413958 A CN 110413958A
- Authority
- CN
- China
- Prior art keywords
- character
- state
- transformation
- automatic machine
- statusline
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/90335—Query processing
- G06F16/90344—Query processing by using string matching techniques
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The present invention provides a kind of linear congruence character set transform method for automatic machine space compression, and step includes: mode trail to be built into automatic machine, and generate state-transition matrix；Each statusline of reading state transfer matrix calculates optimized transformation parameters and maximum effective successor states；According to state-transition matrix and optimized transformation parameters, data structure is recorded, statusline is replaced with into transformed effective status row；The character for reading target text carries out character transformation using linear congruence function, obtains transformed character, successor states are obtained under eligible, realize transformation according to character current state.The present invention also provides a kind of linear congruence character set transformation systems for automatic machine space compression, including rule compiler, transformation parameter generator, statusline converter, comparator, compression automatic machine memory, status register, character set converter and text scanner.
Description
Technical field
The invention belongs to information technology fields, and in particular to a kind of linear congruence character set for automatic machine space compression
Transform method and system.
Background technique
String matching algorithm is a kind of searching algorithm, be widely used in bioinformatics, information retrieval, data compression,
The fields such as network invasion monitoring.One character string is the limited character string being defined on finite alphabet Σ, character string
Matching algorithm searches for some string assemble S={ P on a big character string TiIn all character string PiOccurred position
It sets.T is referred to as text, PiReferred to as pattern string, T and PiIt is all defined on the same alphabet Σ.
In string matching field, automatic machine is a kind of important data structure.For example, 1975 by Aho and
The AC automatic machine algorithm that Corasick is proposed (refers to Efficient String Matching:An Aid to
Bibliographic Search), the KMP algorithm proposed by Knuth, Morris and Pratt for 1977 (refers to Fast
Pattern Matching In Strings), it is calculated by the BOM that Allauzen, Crochemore and Raffinot are proposed within 1999
Many algorithms such as method (referring to Factor Oracle:A New Structure for Pattern Matching) all pass through certainly
Motivation realizes quick string matching.Since the scale of mode trail in most applications is usually larger, the automatic machine of generation
Occupied space is larger, and the resource for also influencing its matching speed, therefore reducing automatic machine occupancy, which just becomes one, is worth asking for research
Topic.
Automatic machine is also known as finite state machine, is a kind of for indicating string assemble and providing string matching function
Data structure.For abstract, the automatic machine in string matching algorithm can be expressed as the matrix A that a size is N × 256N×256,
Wherein N is the status number of automatic machine, and 256 be character set size (1 byte).For current state s and character c is inputted, A [s,
C] indicate the next state arrived at, usually indicated with a nonnegative integer or pointer.A [s, c]=- 1 indicates current state s
There is no successor states for input character c.AN×256In each occupied space of statusline be sizeof (int) × 256, altogether
Occupied space sizeof (int) × 256 × N.Since mode trail S is on a grand scale in numerous applications, corresponding automatic machine shape
State number is more, and the space occupied is very considerable, affects the practicability of the string matching algorithm based on automatic machine, therefore, it is necessary to right
The high-efficiency compression method of automatic machine is studied.
Norton in 2004 is in mono- text of Optimizing Pattern Matching for Intrusion Detection
In propose it is a kind of be known as Banded-Row automatic machine compression method.Due in string matching algorithm, the major part of automatic machine
The usually only seldom several successor states of state, for each statusline A [s], directly with sizeof (int) × 256 come
Expression is to waste very much memory space.In order to compress the memory space of AC automatic machine, two integer lb of Banded-RowsWith
ubsRespectively in recording status row A [s] first and the last one successor states transfer character, it may be assumed that
Remove the idle running of A [s] end to end to move, sizeof (int) × (ub is used only in every a lines-lbs+ 3) memory space comes
It indicates, had both remained the random-access characteristic of array, while also saving memory space compared to matrix representation.
2018 Nian Liuyan soldiers etc. propose the automatic machine space compression method based on character set transformation, due to calculating in String matching
In method, state of automata row is usually very sparse, lbsAnd ubsBetween still have a large amount of invalid states, based on character set transformation
Automatic machine space compression method passes through exclusive or functionInput character is converted, and is defined:
Under the action of suitable transformation parameter X [s], reduceWithBetween invalid state number, further save
Memory space is saved.
Existing technical solution is mainly Banded-Row method and the automatic machine space compression method based on character set transformation,
Both schemes all some shortcomings on room and time.Spatially, Banded-Row method and based on character set transformation from
Motivation space compression method is respectively necessary for occupyingWithSpace, work as ubs-lbsOrWhen larger, even if the successor states number of every row
Less, the space occupied remains on considerable.For example, if lbs=0, ubs=255, even if successor states there are two A [s],
Banded-Row method also can not compression space, in another example, whenAnd when A [s, 125] ≠ -1, although A
[s] only there are three successor states, based on the automatic machine space compression method of character set transformation there is still a need for consumption sizeof (int) ×
130 space.On time, since both methods requires check whether character is more than [lb in scannings,ubs] orRange affects matching speed.
Summary of the invention
The object of the present invention is to provide a kind of linear congruence character set transform method for automatic machine space compression and it is
System belongs to the automatic machine compression method for string matching, and this method guarantees that the time complexity of state transfer is O (1), together
When the memory space of data structure can be greatly reduced.
To achieve the above object, the present invention adopts the following technical scheme:
A kind of linear congruence character set transform method for automatic machine space compression, comprising the following steps:
Mode trail is built into automatic machine, and generates state-transition matrix；
Optimized transformation parameters and maximum effective successor states are calculated in each statusline of reading state transfer matrix；
According to state-transition matrix and optimized transformation parameters, data structure is obtained, is replaced statusline according to the data structure
It is changed to transformed effective status row；
The character for reading target text carries out character transformation using linear congruence function, obtains according to character current state
Transformed character；
If the character after variation is no more than maximum effective successor states, the final subsequent shape after obtaining character transformation
State realizes transformation.
Further, it according to each statusline of state-transition matrix and candidate transformation parameter, calculates minimum effectively subsequent
Maximum effective successor states are calculated when effectively successor states are zero to the minimum in state, and first group of note makes maximum effective
The smallest transformation parameter of successor states is optimized transformation parameters.
Further, the calculating formula of effective successor states isWherein A [] is state transfer
The statusline of matrix, c are the character of target text, and s is character current state, and i, j are candidate transformation parameter.
Further, candidate transformation parameter i value is from 0 to 127, and j value is from 0 to 255.
Further, transformed effective status behavior < A [s, M [s]], A [s, (N [s]+M [s]) mod256], A [s, (N
[s] × 2+M [s]) mod256] ..., A [s, (N [s] × lc [s]+M [s]) mod256] >, wherein N [s], M [s], lc [s] they are number
According to structure, A [] is the statusline of state-transition matrix, and s is character current state.
Further, N [s]=2k+1, M [s]=m, lc [s] are equal to maximum effective successor states, wherein k, and m is best becomes
Change parameter.
Further, linear congruence function is fs (c)=N [s] × c+M [s], and wherein N [s], M [s] are data structure, c
For the character of target text, s is character current state.
Further, final successor states are A [s, c '], and wherein A [] is the statusline of state-transition matrix, and s is word
Accord with current state, the transformed character of c '.
A kind of linear congruence character set transformation system for automatic machine space compression, comprising:
Rule compiler establishes state of automata transfer figure for reading, interpretive model trail, and generates state transfer
Matrix；
Transformation parameter generator, for generating optimized transformation parameters；
Statusline converter receives optimized transformation parameters for reading state transfer matrix line by line, and carries out to statusline
Transformation；
Comparator updates compression automatic machine memory for deciding whether according to transformation results, and knot is compared in generation
Fruit；
Automatic machine memory is compressed, for reading above-mentioned transformation results according to comparison result, updates storage inside；
Status register, for storing current state；
Character set converter according to the current state of status register storage and compresses certainly for reading text character by character
The corresponding transformation parameter stored in motivation, converts character；
Text scanner, the character sent according to the current state of status register storage, character set converter and pressure
The statusline stored in contracting automatic machine calculates next state and updates status register.
A kind of computer readable storage medium storing computer program, the computer program include instruction, which works as
The server is made to execute each step in the above method when being executed by the processor of server.
The method of the present invention guarantees that the time complexity of state transfer is O (1), the fast speed of matched data, while can be big
The memory space of amplitude reduction data structure.
Detailed description of the invention
Fig. 1 is character set transformation schematic diagram.
Fig. 2 is a kind of linear congruence character set transformation system structure chart for automatic machine space compression.
Fig. 3 is state of automata transfer figure.
Fig. 4 A-4D is the result statistical chart for testing 1-4.
Specific embodiment
To enable features described above and advantage of the invention to be clearer and more comprehensible, special embodiment below, and institute's attached drawing is cooperated to make
Detailed description are as follows.
Provided by the present invention for linear congruence character set transform method (the hereinafter referred to as congruence change of automatic machine space compression
Change method) it is in the automatic machine space compression method (hereinafter referred to as exclusive or converter technique) based on character set transformation for prototype, guarantee shape
The time complexity of state transfer is O (1), while the memory space of data structure can be greatly reduced.
As shown in Figure 1, the main thought of the invention is through a linear congruence function fs(c)=ns×c+msTo word
Symbol collection is converted, so that the effective status of statusline is continuous as far as possible.In figure, A [s] is a state in state-transition matrix
Row, p is deviant of each successor states in statusline, and c is corresponding input character.Figure left side indicates exclusive or converter technique,
In the method, input character c passes through exclusive or functionIt is mapped to deviant p, although there was only 3 in A [s] has
Successor states are imitated, in order to store first to a last effective successor states, after needing to store 7 in the shadow region of left side
It include 4 invalid successor states after state.In congruence transformation method shown on the right side of the figure, input character c passes through linear congruence
Function fs(c)=3 × c+3 is mapped to deviant p, and the deviant of effective successor states is transformed to a more continuous area
Domain, therefore it may only be necessary to which storing 4 successor states in right shade region may include all effective successor states.
The same with exclusive or converter technique, congruence transformation method is divided into initialization and two stages of matching, is described as follows.
Initial phase:
1. pressing matrix representation, mode set of strings is built into automatic machine.
2. each statusline for state-transition matrix calculates optimized transformation parameters: reading each statusline of automatic machine
A [s], candidate transformation parameter i value change to 127, j value from 0 and change to 255 from 0, calculate minimum effective successor statesIf infs,i,j=0, calculate maximum effective successor statesRemember that first group makes sups,i,jThe smallest parameter i, j k, m.
3. storing transformation parameter and compression automatic machine: record data structure N [s]=2k+1, M [s]=m, lc [s]=
sups,k,m, statusline is replaced with transformed effective status row < A [s, M [s]], A [s, (N [s]+M [s]) mod256], A
[s, (N [s] × 2+M [s]) mod256] ..., A [s, (N [s] × lc [s]+M [s]) mod256] >.
So far, the step of initial phase is fully completed.
Matching stage:
Upon a match, automatic machine is turned by four available states of data structure N, M, lc, A that above-mentioned compression method generates
Move formula:
Detailed process is as follows:
1. reading in a character c in text to be scanned, according to current state s, calculate c'=(N [s] × c+M [s])
mod256；
2. if c'≤lc [s], successor states are A [s, c']；
3. otherwise, returning, it fails to match.
So far, the step of matching stage is fully completed.
As shown in Fig. 2, congruence transformation method of the invention is real by the automatic machine space compression system based on character set transformation
It is now as follows:
1) rule compiler reading, interpretive model trail, establish state of automata transfer figure, and generate state transfer square
Battle array；
2) statusline converter reads the state-transition matrix of rule compiler generation line by line, while it is raw to receive transformation parameter
It grows up to be a useful person the transformation parameter transmitted, statusline is converted, and transformed statusline length is sent to comparator；
3) comparator decides whether to update compression automatic machine memory according to transformation results, and comparison result is sent out
It send to compression automatic machine memory；
4) compression automatic machine memory receives that comparator is sent as a result, raw according to comparison result reading state row converter
At transformation results, update storage inside；
5) character set converter reads text character by character, according to the current state of status register storage and compression automatic machine
The corresponding transformation parameter of middle storage, converts character and is sent to text scanner；
6) character and pressure that text scanner is stored according to status register current state, character set converter are sent
The statusline stored in contracting automatic machine calculates next state and updates status register.
It is specifically addressed by the following examples:
For ease of description, character set Σ={ 0,1,2,3,4,5,6,7,8,9, A, B, C, D, E, F }, character set size are enabled
| Σ |=16, text T=5C5F, mode trail S are as follows:
1 Sample Rules of table
Initial phase:
1. interpretive model trail establishes state of automata transfer figure, as shown in Figure 3；According to state transition diagram, state is established
Shift-matrix A [s, c], as shown in table 2, -1 indicates invalid transfer in table, subsequent after the corresponding character of other digital representations receiving
State；
2 state-transition matrix of table
2. reading a line in A, to transformation parameter i=0...7 and j=0...15, calculateIf infs,i,j=0, it calculatesNote is minimum
Sups,i,jFor sup's, k, m are designated as under corresponding, such as A [0], k=1, m=3, sup0,1,3=(2 × 1+1) × 0+
3mod16=3, A [0]=2, -1,3,1 > of <；
3. every a line in couple A executes aforesaid operations, four data structures N, M, lc, A are obtained as shown in table 3, table 4:
3 transformation parameter of table
| s | N[s] | M[s] | lc[s] |
| 0 | 3 | 3 | 3 |
| 1 | 3 | 12 | 2 |
| 2 | 3 | 9 | 1 |
| 3 | 3 | 12 | 1 |
| 4 | 1 | 9 | 0 |
| 5 | 1 | 13 | 0 |
| 6 | 1 | 11 | 0 |
| 7 | 1 | 2 | 0 |
| 8 | 1 | 1 | 0 |
Table 4 compresses automatic machine
| s | 0 | 1 | 2 | 3 |
| 0 | 2 | -1 | 3 | 1 |
| 1 | 4 | -1 | 7 | |
| 2 | 7 | 5 | ||
| 3 | 6 | 8 | ||
| 4 | 7 | |||
| 5 | 8 | |||
| 6 | 8 | |||
| 7 | 9 | |||
| 8 | 9 |
So far, the step of initial phase is fully completed.
Matching stage:
1. reading in the first character 5 in text T, according to current state 0, c'=3 × 5+3mod16=2 is calculated；
2. due to lc [0]=3, thus c'≤lc [0], therefore successor states are A [0, c']=3；
3. repeating aforesaid operations until s=9.
So far, the step of matching stage is fully completed.
The good effect that the present invention obtains:
The present invention has made following reality under 64 Linux 4.19.3 systems of single machine (8GB memory, CPU are Intel i7)
It tests:
Test program generates the mode trail and text to be matched for establishing automatic machine at random；Mode trail size exists
It is determined respectively in each experiment, size text is fixed as 10MB.
Statistical indicator: initialization time initializes time and matching speed used in occupied space, matched data.
Experiment uses exclusive or converter technique and does comparative experiments, and experimental result is as shown in table 5.
In experiment 1, long 16 bytes of pattern string, pattern string is concentrated with 262144 pattern strings, exclusive or converter technique occupied space
EMS memory occupation is down to 245.23MB by 344.30MB, congruence transformation method, and in scanning speed, exclusive or converter technique is 10.661MB/s,
Congruence transformation method is 11.318MB/s, is slightly promoted, such as Fig. 4 A.
In experiment 2, long 16 bytes of pattern string, pattern string is concentrated with 524288 pattern strings, exclusive or converter technique occupied space
EMS memory occupation is down to 415.10MB by 751.55MB, congruence transformation method, and in scanning speed, exclusive or converter technique is 9.406MB/s, together
Remaining converter technique is 10.437MB/s, is slightly promoted, such as Fig. 4 B.
In experiment 3, long 32 bytes of pattern string, pattern string is concentrated with 262144 pattern strings, exclusive or converter technique occupied space
EMS memory occupation is down to 439.35MB by 846.81MB, congruence transformation method, and close to halving, in scanning speed, exclusive or converter technique is
15.314MB/s, congruence transformation method are 18.231MB/s, speed-raising about 19%, such as Fig. 4 C.
In experiment 4, long 32 bytes of pattern string, pattern string is concentrated with 524288 pattern strings, exclusive or converter technique occupied space
EMS memory occupation is down to 920.29MB by 2203.26MB, congruence transformation method, and saving is more than half memory, and in scanning speed, exclusive or becomes
Changing method is 13.114MB/s, and congruence transformation method is 16.751MB/s, speed-raising 27.7%, such as Fig. 4 D.
5 experimental result of table statistics
The above experiment shows that the occupied significant spatial of compression automatic machine of the method for the present invention converts compression side lower than exclusive or
The speed of method, matched data is faster than exclusive or transform-based image compression, achieves apparent technical effect.Therefore, this method and system
There are extensive real value and application scenarios.
The above embodiments are merely illustrative of the technical solutions of the present invention rather than is limited, the ordinary skill of this field
Personnel can be with modification or equivalent replacement of the technical solution of the present invention are made, without departing from the spirit and scope of the present invention, this
The protection scope of invention should be subject to described in claims.
Claims (10)
1. a kind of linear congruence character set transform method for automatic machine space compression, which comprises the following steps:
Mode trail is built into automatic machine, and generates state-transition matrix；
Optimized transformation parameters and maximum effective successor states are calculated in each statusline of reading state transfer matrix；
According to state-transition matrix and optimized transformation parameters, data structure is obtained, is replaced with statusline according to the data structure
Transformed effective status row；
The character for reading target text carries out character transformation using linear congruence function, is converted according to character current state
Character afterwards；
If the character after variation is no more than maximum effective successor states, the final successor states after obtaining character transformation are real
Now convert.
2. the method as described in claim 1, which is characterized in that according to each statusline and candidate transformation of state-transition matrix
Parameter calculates minimum effective successor states, and when effectively successor states are zero to the minimum, maximum effectively subsequent shape is calculated
State, first group of note make the maximum effectively the smallest transformation parameter of successor states be optimized transformation parameters.
3. method according to claim 2, which is characterized in that the meter in of effective successor states calculates fs, i formula, j isWherein A [] is the statusline of state-transition matrix, and c is the character of target text, and s is character
Current state, i, j are candidate transformation parameter.
4. method as claimed in claim 3, which is characterized in that candidate transformation parameter i value is from 0 to 127, and j value is from 0
To 255.
5. the method as described in claim 1, which is characterized in that transformed effective status behavior < A [s, M [s]], A [s, (N
[s]+M [s]) mod256], A [s, (N [s] × 2+M [s]) mod256] ..., A [s, (N [s] × lc [s]+M [s]) mod256] >,
Wherein N [s], M [s], lc [s] are data structure, and A [] is the statusline of state-transition matrix, and s is character current state.
6. method as claimed in claim 5, which is characterized in that N [s]=2k+1, M [s]=m, after lc [s] is equal to maximum effectively
After state, wherein k, m are optimized transformation parameters.
7. the method as described in claim 1, which is characterized in that linear congruence function is fs (c)=N [s] × c+M [s], wherein
N [s], M [s] are data structure, and c is the character of target text, and s is character current state.
8. the method as described in claim 1, which is characterized in that final successor states are A [s, c '], and wherein A [] is state
The statusline of transfer matrix, s are character current state, the transformed character of c '.
9. a kind of linear congruence character set transformation system for automatic machine space compression characterized by comprising
Rule compiler establishes state of automata transfer figure, and generate state-transition matrix for reading, interpretive model trail；
Transformation parameter generator, for generating optimized transformation parameters；
Statusline converter receives optimized transformation parameters, and become to statusline for reading state transfer matrix line by line
It changes；
Comparator updates compression automatic machine memory for deciding whether according to transformation results, generates comparison result；
Automatic machine memory is compressed, for reading above-mentioned transformation results according to comparison result, updates storage inside；
Status register, for storing current state；
Character set converter, for reading text character by character, according to the current state of status register storage and compression automatic machine
The corresponding transformation parameter of middle storage, converts character；
Text scanner, the character sent according to the current state of status register storage, character set converter and compression are certainly
The statusline stored in motivation calculates next state and updates status register.
10. a kind of computer readable storage medium for storing computer program, which is characterized in that the computer program includes referring to
It enables, which makes the server execute any side the claims 1-8 when the processor execution by server
Each step in method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910505446.2A CN110413958B (en) | 2019-06-12 | 2019-06-12 | Linear congruence character set transformation method and system for automaton space compression |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910505446.2A CN110413958B (en) | 2019-06-12 | 2019-06-12 | Linear congruence character set transformation method and system for automaton space compression |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110413958A true CN110413958A (en) | 2019-11-05 |
| CN110413958B CN110413958B (en) | 2020-12-04 |
Family
ID=68358997
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910505446.2A Active CN110413958B (en) | 2019-06-12 | 2019-06-12 | Linear congruence character set transformation method and system for automaton space compression |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110413958B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117675417A (en) * | 2024-02-02 | 2024-03-08 | 中国电子信息产业集团有限公司第六研究所 | Quick text scanning method and device, electronic equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004013777A1 (en) * | 2002-08-05 | 2004-02-12 | Fish Robert | System and method of parallel pattern matching |
| CN101630323A (en) * | 2009-08-20 | 2010-01-20 | 中国科学院计算技术研究所 | Method for compressing space of finite automaton |
| CN101916259A (en) * | 2010-07-06 | 2010-12-15 | 中国科学院计算技术研究所 | Space compression method of state transition table of deterministic automaton |
| CN103559018A (en) * | 2013-10-23 | 2014-02-05 | 东软集团股份有限公司 | String matching method and system based on graphics processing unit (GPU) calculation |
| US9083740B1 (en) * | 2009-09-28 | 2015-07-14 | Juniper Networks, Inc. | Network traffic pattern matching using adaptive deterministic finite automata |
| CN104809161A (en) * | 2015-04-01 | 2015-07-29 | 中国科学院信息工程研究所 | Method and system for conducting compression and query on sparse matrix |
| CN104881439A (en) * | 2015-05-11 | 2015-09-02 | 中国科学院信息工程研究所 | Method and system for space-efficient multi-pattern matching |
| CN105426412A (en) * | 2015-11-03 | 2016-03-23 | 北京锐安科技有限公司 | Multi-mode string matching method and device |
-
2019
- 2019-06-12 CN CN201910505446.2A patent/CN110413958B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004013777A1 (en) * | 2002-08-05 | 2004-02-12 | Fish Robert | System and method of parallel pattern matching |
| CN101630323A (en) * | 2009-08-20 | 2010-01-20 | 中国科学院计算技术研究所 | Method for compressing space of finite automaton |
| US9083740B1 (en) * | 2009-09-28 | 2015-07-14 | Juniper Networks, Inc. | Network traffic pattern matching using adaptive deterministic finite automata |
| CN101916259A (en) * | 2010-07-06 | 2010-12-15 | 中国科学院计算技术研究所 | Space compression method of state transition table of deterministic automaton |
| CN103559018A (en) * | 2013-10-23 | 2014-02-05 | 东软集团股份有限公司 | String matching method and system based on graphics processing unit (GPU) calculation |
| CN104809161A (en) * | 2015-04-01 | 2015-07-29 | 中国科学院信息工程研究所 | Method and system for conducting compression and query on sparse matrix |
| CN104881439A (en) * | 2015-05-11 | 2015-09-02 | 中国科学院信息工程研究所 | Method and system for space-efficient multi-pattern matching |
| CN105426412A (en) * | 2015-11-03 | 2016-03-23 | 北京锐安科技有限公司 | Multi-mode string matching method and device |
Non-Patent Citations (1)
| Title |
|---|
| 熊刚、何慧敏、于静、刘燕兵、郭莉: "HybridFA:一种基于统计的AC自动机空间优化技术", 《通信学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117675417A (en) * | 2024-02-02 | 2024-03-08 | 中国电子信息产业集团有限公司第六研究所 | Quick text scanning method and device, electronic equipment and storage medium |
| CN117675417B (en) * | 2024-02-02 | 2024-04-16 | 中国电子信息产业集团有限公司第六研究所 | Quick text scanning method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110413958B (en) | 2020-12-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108959246B (en) | Answer selection method and device based on improved attention mechanism and electronic equipment | |
| Ren et al. | On querying historical evolving graph sequences | |
| Chen et al. | Density-based clustering for real-time stream data | |
| US20130198272A1 (en) | Operation log storage system, device, and program | |
| CN103593440A (en) | Method and device for reading and writing log file | |
| CN112884204B (en) | Network security risk event prediction method and device | |
| CN112231514B (en) | Data deduplication method and device, storage medium and server | |
| CN101604408B (en) | Generation of detectors and detecting method | |
| CN110401451A (en) | Automatic machine space compression method and system based on character set transformation | |
| CN110389840B (en) | Load consumption early warning method and device, computer equipment and storage medium | |
| Yong et al. | Efficient graph summarization using weighted lsh at billion-scale | |
| CN110413958A (en) | Linear congruence character set transform method and system for automatic machine space compression | |
| CN113761192B (en) | Text processing method, text processing device and text processing equipment | |
| Zhang et al. | SPOT: A system for detecting projected outliers from high-dimensional data streams | |
| CN111767419B (en) | Picture searching method, device, equipment and computer readable storage medium | |
| CN112181302A (en) | Data multilevel storage and access method and system | |
| CN116127447A (en) | Virtual power plant false data injection attack detection method, device, terminal and medium | |
| CN108399152A (en) | Compression expression method, system, storage medium and the rule match device of digital search tree | |
| CN112054805B (en) | Model data compression method, system and related equipment | |
| KR20180137387A (en) | Apparatus and method for detecting overlapping community | |
| Liu et al. | An analysis of missing data treatment methods and their application to health care dataset | |
| CN113342518A (en) | Task processing method and device | |
| CN113468202B (en) | Memory data screening method, device, equipment and storage medium | |
| CN115329118B (en) | Image similarity retrieval method and system for garbage image | |
| JPWO2020074788A5 (en) |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |