CN104361111B - A kind of archives are compiled and grind method automatically - Google Patents
A kind of archives are compiled and grind method automatically Download PDFInfo
- Publication number
- CN104361111B CN104361111B CN201410714594.2A CN201410714594A CN104361111B CN 104361111 B CN104361111 B CN 104361111B CN 201410714594 A CN201410714594 A CN 201410714594A CN 104361111 B CN104361111 B CN 104361111B
- Authority
- CN
- China
- Prior art keywords
- mrow
- volume
- archives
- expert
- classification
- 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.)
- Active
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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3322—Query formulation using system suggestions
- G06F16/3323—Query formulation using system suggestions using document space presentation or visualization, e.g. category, hierarchy or range presentation and selection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to the classification of documents and retrieval technique field, being related to a kind of archives based on B/S framework MIS of Department Files, volume grinds method automatically；It is first according to the unified form difference typing archive information that archives list, list in volume and expert's registration card interface are provided, archive information is classified and collected automatically using automatic hierarchical classification algorithm with management module by archives typing again, and is stored in respectively in corresponding database；Then the compiling and researching condition and the information of deposit that Compiling of Files module is inputted according to user are retrieved to associated databases, inquired about and collected generation Compiling of Files result, finally Compiling of Files result is included on screen, or preserve Compiling of Files result to form paper document after export printing in the form of Word document or Excel forms, realize that the automatic volume of archives is ground；Its design principle science is reliable, and volume grinds that labor intensity is small, and operating efficiency is high, and information careless omission is few, it is ensured that compile the quality and value ground, and volume grinds efficiency high, and volume grinds environment-friendly.
Description
Technical field：
The invention belongs to the classification of documents and retrieval technique field, it is related to a kind of based on B/S framework MIS of Department Files
Archives compile automatically and grind method, be that archives compilation and archives digest compilation provide the automatic volume of the archives completed by computer software and ground
Technology.
Background technology：
MIS of Department Files based on B/S frameworks is extension of the Traditional Archives function in informationized society, is both had
There are the base attribute and function of Traditional Archives, the development need of information age is adapted to again, realizing conventional archival digitalization
While management, realized by internet with digital archives data storage storehouse is set up to all departments and the receipts of all kinds of archive informations
Collection, storage, management and use, Informatization Service is provided for the utilization of Archive Resource.Compiling of Files towards File use works
It is actual demand of the archives/room according to File use, based on archives/room stock archives, with assemble archive with reference to money
Expect a specialized work for achievement form.The essence of Compiling of Files work is that the content of files is studied and processed
Arrange, and volume is ground into volume, is allowed to very clear, to improve department in power and unit integrated management level and operating efficiency, enhancing
Value of services of the Archive Resource to society.At present, based on Compiling of Files work is mainly ground with artificial volume, its speed is slow, efficiency
Low, volume grinds of poor quality.
Traditional artificial method of compiling and researching archives is divided into 2 kinds according to the processing level to archives, the first Compiling of Files plus
Work method is to take passages, reduce the staff and editing on the basis of raw file, forms summary material；It processes the achievement of Compiling of Files
Form has：Dispatch collects, special topic compilation, records series digest compilation, experts and scholars and academic opinion including a certain field
Digest compilation, scientific and technological achievement digest compilation etc.；The processing method of second of Compiling of Files is to need to provide to raw file
On the basis of the relevant content of material is analyzed, studied and concluded, new material is write out；It processes Compiling of Files achievement form
Have：Yearbook, evolution of organization, Shi Zhi and the report of integrated technology economic research etc..Fruit shape is processed into second Compiling of Files
Due to containing people to new contents such as the new knowledge of things, new viewpoint, new conclusion and new suggestions in formula, from information content
These are all the information newly increased from the point of view of angle, so the processing of this Compiling of Files is usually the expert or scholar by association area
Complete.And the information that the first processing Compiling of Files achievement is included is existing information in archives, information content will not be increased, do not produced
Raw new content, makes every effort to accomplish " entirely, essence, standard ", it is impossible to have careless omission, the content that volume is ground is comprehensive, careful, peaceful not leak more, and
Continuity over time will reach mass data amount, it is necessary to compile the raw file material accumulation ground, for mass data by people
Work progress volume grinds slightly general idea and will result in information omission or malfunction, it is impossible to ensure the quality and value of Compiling of Files achievement, its
Labor intensity is big, inefficiency, and it is poor that volume grinds accuracy rate, constrains Archive Resource in mass data scale and more high-tech level
On abundant development and utilization, human cost is very high.
The content of the invention：
It is an object of the invention to overcome the shortcoming that prior art is present, seek design and provide a kind of based on file administration letter
The archives of breath system are compiled and grind method automatically, and Compiling of Files is automatically generated into fruit shape using computer automatic sorting and retrieval technique
Formula, improves Compiling of Files efficiency and accuracy rate, reduces the artificial information careless omission compiled when grinding.
To achieve these goals, the present invention in the MIS of Department Files based on B/S frameworks by archives typing with
Management module and Compiling of Files module, which are combined, realizes that archives automatically grind by volume, and it is concretely comprised the following steps：
(1) archive information typing, the archives list shown according to system, list in volume and expert's registration card interface are first carried out
Unified form difference typing archives title, the affiliated classification of archives, shelves number, year and all kinds of essential informations of expert's situation provided；
(2) it is automatic to step using automatic hierarchical classification algorithm proposed by the present invention by archives typing and management module again
(1) archive information of typing is classified and collected, and it is basic to be stored in corresponding archives catalog, list in volume and expert respectively
Information is registered in catalog data base and expert database；
(3) compiling and researching condition that is inputted again by Compiling of Files module according to user and the information of deposit to corresponding archives catalog,
List in volume registers catalog data base with expert's essential information and expert database is retrieved, inquired about and collected generation archives and compiled
Grind result；
(4) Compiling of Files result is included on screen, or by Compiling of Files result with Word document or Excel forms
Printout forms paper document preservation after form export, realizes that the automatic volume of archives is ground.
Automatic hierarchical classification algorithm proposed by the present invention is the improvement to existing conventional NB Algorithm, simple shellfish
This algorithm of leaf refers to consider that all features of text classify to text during classification, ties forecast sample according to prediction during classification
Fruit is divided into particular document class probability highest class library.
The specific disaggregated model of NB Algorithm of the present invention is as follows：The archives text of a given unknown classification
This X, provided with m classification, is designated as C1, C2 ... ..., Cm, according to Naive Bayes Classification law, has under condition X after highest
The calculation formula for testing the classification P (Ci | X) of probability is as follows：
In P (Ci | X) calculation formula, P (X) is constant, therefore only needs to maximize molecule P (X | Ci) P (Ci) i.e.
Can；P (Ci) is the category distribution probability in training set, and calculation formula is：Molecule is classification in formula | Ci |
Comprising textual data plus 1, denominator is m classification and | D | be the text sum sum included in training set；In order to simplify P (X |
Ci calculating process), it is assumed that multiple attributes of text are independent of each other, therefore, and calculating P (X | Ci) it is exactly to calculate characteristic attribute
The probability occurred on classification Ci, P (X | Ci) value is calculated using 2 kinds of computation models of Laplace estimation：
(1) whether multivariate model, statistical nature attribute occurred in the text, if occurring being designated as 1, was otherwise designated as 0.Meter
Calculating formula is：
Wherein, | V | the total quantity of representation eigenvalue, BxtIt is wtThe mark occurred in text X, if wtThere is then BxtIt is designated as
1, otherwise it is designated as 0, wtRepresent t-th of feature, i.e. t-th vectorial of component, therefore, the P (w in formulat|Ci) calculation formula is as follows：
(2) multinomial model (Multinomial Model) the then occurrence number of statistical nature attribute in the text, is calculated
Formula is：
Wherein, NxtRepresent the number of times that feature t occurs in text X；P(wt|Ci) calculation formula it is as follows：
In P (wt|Ci) calculation formula in, NjtT is characterized in text djIn the number of times that once occurred, | D | it is total for training text
Number, | V | it is characterized sum, NjsIt is feature s in text djIn occurrence number；The essence of the sorting technique is in text object
All characteristic values probability for being counted and being mapped in each already present classification.
The present invention is improved NB Algorithm in archives typing management module, is realized and is based on archives list
The automatic hierarchical classification algorithm of title and keyword rough sort, directly extracts crucial from the topic of archives list and list in volume
Word set, builds stratification disaggregated model, and classifying quality is reached with low characteristic dimension after appropriate dimensionality reduction, replaces traditional text
The Chinese word segmentation of sorting algorithm, effectively improves the nicety of grading and operational efficiency of file literature；It is described based on archives list mark
The implementation process of the automatic hierarchical classification algorithm of topic and keyword rough sort is as follows：
(1) first in local or online input system archive information, the archives list shown according to system, list in volume and expert step on
All kinds of basic letters of unified form difference typing archives title, affiliated classification, shelves number, year and expert that note card interface is provided
Breath；
(2) system automatically extracts the text data characteristic parameter collection of archives title and the keyword in archives text and preserved
In corresponding database；
(3) dimensionality reduction is carried out when exceeding threshold value to the text data characteristic parameter collection of extraction, excessive feature frequently can lead to
Dimension disaster, making the efficiency of classification reduces；
(4) rough sort of Naive Bayes Classification Algorithm is performed according to the text data characteristic parameter or keyword of extraction；
(5) feature extraction is carried out for each subclass respectively again in step (4) rough sort result；
(6) the text data characteristic parameter execution Naive Bayes Classification Algorithm again for each subclass is automatically performed subdivision
Class；
(7) output category result and it is saved in corresponding database.
The data message of Compiling of Files resume module of the present invention is for the archive information typing set up and pipe
The basic volume for managing archives list, list in volume and expert's registration card catalogue progress of module is ground, in archive information typing and management
Being created in module includes archives list database, list in volume database, classification of documents database, expert's essential information registrating number
According to 6 databases such as storehouse, the civilized thin and expert's Item Details of Specialists；Compiling of Files module by the classification of documents compile grind submodule,
File font size index volume grinds submodule and expert info is compiled and grinds 3 submodule compositions, is created and above-mentioned 6 in Compiling of Files module
The associated Compiling of Files basic database of individual database；The classification of documents is compiled and grinds the compiling and researching condition that submodule is inputted according to user,
It is automatic to realize that classification of documents volume is ground, archives title is compiled to grind to compile with time of filing and grinds and show that volume grinds result with tabular form；File
Font size index, which is compiled to grind submodule and grind demand according to the volume of user and input, includes official document font size, year, shelves number and the group of retention period
Compiling and researching condition is closed, is clicked on according to condition filter data after inquiry, to automatically generate and show that file font size index volume grinds list；
Expert info volume grinds submodule and grinds statistics to expert's registration card information progress volume in file administration, according to expert's surname of input
The compiling and researching condition of name, research direction and achievement carries out fuzzy query, realizes that expert classification volume is ground, expert's research direction is compiled and ground, specially
Family's paper information is compiled to grind to compile with expert's project information and ground, and volume is ground after result is collected including on screen；Wherein expert
Classification, which compiles to grind to refer to compile to grind all expert infos for automatically generating a certain research field and compile, grinds the results list, and it is compiled and grinds result and can lead
Go out into Excel or Word document to preserve or print out.
The Compiling of Files that classification of documents volume of the present invention grinds submodule comprises the following steps：
(1) create classification volume and grind view, view is using list in volume or archives list as main table, and associative classification information table is obtained
Classification information title, list in volume is stored in from archives list information in different tables of data, needed when classification volume is ground
Collect two parts information, carry out unified query and retrieve；
(2) data access layer identification code, retrieval needs to compile the archives ground during data access layer is the view provided from step (1)
Information, performs function using querying condition as parameter, retrieves qualified archive information, and classification volume grinds the extraction classification of documents, shelves
Case title, time of filing, number, number of pages and shelves information, and when result is ground to volume according to specific name, distribution caption and filing
Between be ranked up；
(3) classification volume grinds application layer realization, classification compiling and researching condition is set first, by the fuzzy search of classification of documents title, shelves
The fuzzy search of case title and time of filing searching classification, which are compiled, to be ground；
(4) volume is ground into result to export in EXCEL or Word document, facilitates user's preservation to check and print and bind into book form.
The Compiling of Files that file font size index volume of the present invention grinds submodule comprises the following steps：
(1) establishment file font size indexed view, file font size indexed view is using list in volume as main table associative classification information
Table, extracts official document font size, year, file sequence number, shelves number, page number, number of pages and retention period information；
(2) data access layer identification code, data access layer is to extract information from file font size indexed view, and according to official document
Font size, file sequence number and year sequence, function is using querying condition as parameter, by application layer dynamic construction；
(3) application layer main code, default compiling and researching condition, including official document font size, year, shelves number and retention period,
Needs are ground according to volume and are combined input, are clicked on after inquiry according to condition filter data；
(4) volume is ground into result to export in EXCEL or Word document, facilitates user's preservation to check and print and bind into book form.
The Compiling of Files that expert info volume of the present invention grinds submodule comprises the following steps：
(1) create expert info volume and grind view, view is using expert's essential information registration form as main table, association Specialists civilization
Thin table extracts paper information, and association expert item details extract project and winning information, the registration of expert's essential information and expert
Achievement information is stored in different tables of data, is needed to collect each several part information when row information is compiled and ground entering, is carried out unified query；
(2) data access layer identification code, data access layer is to obtain to need to compile the expert's letter ground from the view in step (1)
Breath, function inquires about qualified expert's archive information using querying condition as parameter, volume grind extraction expert name, expert's classification,
Research direction, paper information and project information；
(3) expert info, which is compiled, grinds application layer realization, sets compiling and researching condition, carries out the fuzzy search of expert's title, expert's research
Direction fuzzy search, thesis topic summarizes fuzzy search, the retrieval of paper publishing time, project name fuzzy search, item with paper
Mesh summarizes fuzzy search and the retrieval of project beginning and ending time and the prize-winning situation retrieval of project；Expert info is compiled to grind and ground according to the volume of input
Condition, realizes that expert classification volume is ground, expert's research direction volume is ground, expert's paper information is compiled to grind to compile with expert's project information and ground；
(4) volume is ground into result to export in Excel or Word document, facilitates user's preservation to check and print and bind into book form.
The execution flow of MIS of Department Files of the present invention is：
(1) a browser is opened in client, the station address of input system is sent out to Web server in address field
Go out service request, when showing the login page of system on desktop, after login page fills in user name, password and identifying code
Send Web server to, Web server uses http protocol the master of MIS of Department Files after being verified to user identity
Page sends client to, and client browser receives the homepage file transmitted, and it is shown on screen；
(2) typing of archives essential information, the archives list, list in volume and expert for being shown in homepage according to system is registered
Block classification, shelves number, archives title, year and all kinds of basic letters of expert belonging to the unified form difference typing provided and addition archives
Breath；System performs corresponding extension application in the Business Logic of Web server and is attached with database server, leads to
Cross SQL modes the above-mentioned all kinds of essential informations of user's typing or addition are connected in storage to corresponding with Web server
Before database, catalogue and archives title are carried out automatically classifying and filing by system, then original text is mounted；Original text can be electricity
Sub- scanned copy or electronic edition original paper；
(3) when needing to carry out Compiling of Files to certain class archive information, the entry for the Compiling of Files that system is selected according to user,
Corresponding Compiling of Files interface is entered, compiling and researching condition is inputted in the interface；For example：Carry out the expert info in some field
When volume is ground, then the title or research direction for inputting the research field also grind parameter click inquiry including the volume such as paper, project situation
Associated databases are carried out information retrieval and inquiry by button；
(4) volume inputted according to (3) grinds parameter, performed with Web server link after by SQL statement to corresponding
Database server propose data processing request, i.e., to Compiling of Files basic database and be associated other database informations
Retrieved and inquiry operation, and the data item that meets compiling and researching condition will be retrieved and counted, analyzed and collected, generate archives
Volume is ground into fruit；
(5) result of the Compiling of Files of generation is submitted to Web server by database server, then is passed by Web server
It is sent to client and is shown on screen；
(6) volume is ground into fruit and exports to preservation or printout in Word document or Excel forms.
Compared with prior art, its design principle science is reliable by the present invention, and volume grinds that labor intensity is small, and operating efficiency is high, letter
Breath careless omission is few, it is ensured that compile the quality and value ground, and volume grinds efficiency high, and volume grinds environment-friendly.
Brief description of the drawings：
Fig. 1 constitutes structural principle schematic block diagram for the hardware of apparatus of the present invention.
Fig. 2 is the logical functional structure principle schematic block of Compiling of Files module of the present invention and module for managing files
Figure.
Automatic compile of MIS of Department Files that Fig. 3 is the present invention grinds execution flow chart.
Fig. 4 is the hierarchical classification algorithm execution flow chart of the rough sort of the present invention.
Embodiment：
It is described further below by embodiment and with reference to accompanying drawing.
Embodiment 1：
The present embodiment carries out test and evaluation to sorting algorithm proposed by the present invention, first in 1000 archives texts being collected into
In this, classification based training, remaining 960 archives text are carried out to method of the present invention from every 40 texts of class random selection
The test evaluation of classification results is just carried out to method of the present invention as text set to be sorted；Wherein, secretarial document class
It it is 222, scientific and technical archive class 216, accounting file class 162, personal file class 95, sound and video archive class 43 is comprehensive to shine
Piece class 86, archival objects class 35, archive file class 40, periodical archives class is 61, respectively with precision ratio, recall ratio
Classification results are evaluated with F1 (harmonic-mean of recall ratio and precision ratio) three indexs of test value, test evaluation result
As shown in table 1；
The classification results test evaluation table of table 1
| Classification | Recall ratio | Precision ratio | F1 test values |
| Secretarial document class | 95.08% | 91.80% | 93.44% |
| Scientific and technical archive class | 85.34% | 87.93% | 86.64% |
| Accounting file class | 90.32% | 93.55% | 91.94% |
| Personal file class | 92.00% | 94.67% | 93.33% |
| Sound and video archive class | 93.02% | 95.35% | 94.19% |
| Photomonate class | 97.67% | 94.19% | 95.93% |
| Archival objects class | 91.43% | 94.29% | 92.86% |
| Archive file class | 87.50% | 85.00% | 86.25% |
| Periodical archives class | 90.16% | 83.61% | 86.89% |
Upper table illustrates that recall ratio, precision ratio and the F1 test values of the classification results of the present embodiment can reach preferably
Effect, the intrinsic dimensionality produced during rough sort according to Document Title and keyword improves system fortune below 50
Line efficiency.
The running environment requirement of the present embodiment：Configure dual microprocessors or higher, more than internal memory 2G networking PC and simultaneous
Appearance machine is run；Server OS is Windows XP and above version；VS .NET Prerequisites software is Framework 3.5, SQL
Server2005；Exploitation software is Microsoft Visual Studio 2008；Using B S three-tier architectures, pass through asp.net
Expression layer, Business Logic and data layer identification code are realized respectively.
It is server addition use current embodiment require that installing and setting Microsoft SQL Server database servers
Import system database after name in an account book and setting password；Then issuing web site (i.e. the MIS of Department Files of B/S frameworks)；Net
Stand after issuing successfully, open any one browser networked on PC, input station address, which is entered, in address field steps on
Click logs into system administration main interface after recording the page, input account, password and identifying code；On the left of system manager
Tree menu column in click on【Compiling of Files】, showing on screen is needed to compile the multiple entries ground, including classification volume is ground, file word
Number index volume grind with expert info compile grind；Select and click on【Classification volume is ground】, enter classification volume and grind interface, in the interface
Each compiling and researching condition is inputted, including classification of documents title, archives title and time of filing (are more than sometime, less than sometime
Or in certain time) etc., system can automatically generate classification according to the compiling and researching condition of input and compile the result ground and in the form of a list
It is shown on screen, and preservation or printing in Word document or Excel can be exported to；Select and click on【File font size index is compiled
Grind】, then enter file font size index volume and grind interface, being ground in the interface according to volume needs to be combined the content that input volume is ground
Such as official document font size, year, shelves number and retention period, system after inquiry of clicking on can be according to the retrieval of the compiling and researching condition of input, filtering
Data automatically generate and show that volume grinds result, or volume is ground into result exported in Word document or Excel and preserve or print；Selection
And click on【Expert info is compiled and ground】, enter expert info volume and grind interface, compiling and researching condition is inputted in the interface, system carries out special
Family's name fuzzy search, research direction fuzzy search, thesis topic, paper general introduction fuzzy search, the retrieval of paper publishing time, item
Whether the fuzzy search of mesh title, project overview fuzzy search, the retrieval of project beginning and ending time and the prize-winning situation of project are (prize-winning, prize-winning
Title) retrieval；System is according to expert's name of input, and the compiling and researching condition of research direction and achievement carries out fuzzy query, realizes special
Family's classification volume is ground, expert's research direction volume is ground, expert's paper information is compiled to grind to compile with expert's project information and ground；And volume is ground into result entered
Row is shown on screen after collecting, or result is exported into preservation or printout in Excel or Word document.
The classification of documents management of the present embodiment carries out the setting that classification volume is ground, actual work according to the actual demand of Compiling of Files
Volume is needed to grind the classification of documents of one " administrative examination and approval ", system manager only needs to increase in classification of documents table one " OK
The grouped data of political examination batch ", you can safeguard the archive information accordingly classified in the list in volume, archives list in archives, and it is right
The classification performs volume and grinds operation；System is collected all kinds of archive informations by the typing to all kinds of archive informations, according to point
The relevant information quick-searching such as class or archives title goes out the archives, by checking the inventory locations of archives, quickly from corresponding thing
Reason deposit position obtains the archives, shortens the time that archives are found, and improves operating efficiency.
Embodiment 2：
The automatic device for compiling method of grinding of archives that the present embodiment is related to is realized, its agent structure is by client browser 1, shelves
Case Data Enter connects composition with management module 2, Compiling of Files module 3 and with 7 database power information that it is linked.Its
In, archive information typing is with management module 2 again including list in volume management submodule 4, archives list management submodule 5 and expert
6 three functional module elements of registration card message sub-module；Compiling of Files module 3 is compiled by the classification of documents and grinds submodule 7, file font size
Index volume, which is ground submodule 8 and expert info and compiled, grinds 9 three functional module elements power information of submodule and connects composition；Archives typing with
The typing archive information of management module 2 is simultaneously safeguarded to archive information, to the shelves of list in volume, archives list and expert's registration card
Maintenance is classified and collected to case information；The compiling and researching condition that Compiling of Files module 3 is inputted according to user, realizes the classification of documents automatically
Volume is ground, archives title is compiled to grind to compile with time of filing and ground, and shows that volume grinds result with tabular form；File font size index volume grinds submodule
Block 8 grinds the compiling and researching condition that demand input is combined by official document font size, year, shelves number and retention period according to the volume of user, and click is looked into
Filter data can be crossed according to compiling and researching condition after inquiry, to automatically generate and show that file font size index volume grinds list；Expert info
Volume grinds submodule 9 and grinds statistics to expert's registration card information progress volume in file administration, according to expert's name of input, research side
Fuzzy query is carried out to the compiling and researching condition with achievement, realizes that expert classification volume is ground, expert's research direction volume is ground, expert's paper information
Volume is ground to compile with expert's project information and ground, and volume is ground after result is collected including on screen；Wherein expert classification, which is compiled to grind, is
Compile to grind all expert infos for automatically generating a certain research field and compile and grind the results list, it is compiled and grinds result and can export to
Preserve or print out in Excel or Word document.With archive information typing and management module 2 and the telecommunications of Compiling of Files module 3
7 databases of manner of breathing link are respectively archives list database, list in volume database, classification of documents database, Zhuan Jiaji
The literary detailed data storehouse of this information registered database, Specialists, expert's Item Detail database and Compiling of Files basic database；Visitor
Family end browser 1 is any browser software run on any computer and terminal device of networking.
Claims (6)
1. a kind of archives are compiled and grind method automatically, it is characterised in that by archives in the MIS of Department Files based on B/S frameworks
Typing is combined with management module and Compiling of Files module realizes that archives automatically grind by volume, and it is concretely comprised the following steps：
(1) archive information typing is first carried out, the archives list shown according to system, list in volume and expert's registration card interface are provided
Unified form difference typing archives title, classification, shelves number, annual and all kinds of essential informations of expert's situation belonging to archives；
(2) step (1) is recorded using automatic hierarchical classification algorithm proposed by the present invention is automatic by archives typing and management module again
The archive information entered is classified and collected automatically, and it is basic to be stored in corresponding archives catalog, list in volume and expert respectively
Information is registered in catalog data base and expert database；
(3) compiling and researching condition and the information of deposit inputted again by Compiling of Files module according to user is in corresponding archives catalog, volume
Catalogue registers catalog data base with expert's essential information and expert database is retrieved, inquired about and collected and generates Compiling of Files knot
Really；
(4) Compiling of Files result is included on screen, or by Compiling of Files result in the form of Word document or Excel forms
Printout forms paper document preservation after export, realizes that the automatic volume of archives is ground；
The implementation process of described automatic hierarchical classification algorithm is as follows：
(1) first in local or online input system archive information, the archives list shown according to system, list in volume and expert's registration card
All kinds of essential informations of unified form difference typing archives title, affiliated classification, shelves number, year and expert that interface is provided；
(2) system automatically extracts the text data characteristic parameter collection of archives title and the keyword in archives text and is stored in phase
In the database answered；
(3) dimensionality reduction is carried out when exceeding threshold value to the text data characteristic parameter collection of extraction, excessive feature frequently can lead to dimension
Disaster, making the efficiency of classification reduces；
(4) rough sort of Naive Bayes Classification Algorithm is performed according to the text data characteristic parameter or keyword of extraction；
(5) feature extraction is carried out for each subclass respectively again in step (4) rough sort result；
(6) the text data characteristic parameter execution Naive Bayes Classification Algorithm again for each subclass is automatically performed disaggregated classification；
(7) output category result and it is saved in corresponding database；
The NB Algorithm refers to consider that all features of text classify to text during classification, will predicted during classification
Sample is divided into according to predicting the outcome in particular document class probability highest class library, and its specific disaggregated model is as follows：It is given
The archives text X of one unknown classification, provided with m classification, is designated as C1, C2 ... ..., Cm, according to Naive Bayes Classification law,
The calculation formula of the classification P (Ci | X) with highest posterior probability is as follows under condition X：
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>C</mi>
<mi>i</mi>
<mo>|</mo>
<mi>X</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>|</mo>
<mi>C</mi>
<mi>i</mi>
<mo>)</mo>
</mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>C</mi>
<mi>i</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
</mrow>
In P (Ci | X) calculation formula, P (X) is constant, therefore only needs to maximize molecule P (X | Ci) P (Ci)；P
(Ci) it is category distribution probability in training set, calculation formula is：Molecule is classification in formula | Ci | include
Textual data add 1, denominator is m classification and | D | be the text sum sum included in training set；In order to simplify P (X | Ci)
Calculating process, it is assumed that multiple attributes of text are independent of each other, therefore, and it is exactly to calculate that characteristic attribute exists to calculate P (X | Ci)
The probability occurred on classification Ci, P (X | Ci) value is calculated using 2 kinds of computation models of Laplace estimation：
(1) whether multivariate model, statistical nature attribute occurred in the text, if occurring being designated as 1, was otherwise designated as 0, calculated public
Formula is：
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>|</mo>
<msub>
<mi>C</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Pi;</mo>
<mrow>
<mi>t</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>|</mo>
<mi>v</mi>
<mo>|</mo>
</mrow>
</munderover>
<mrow>
<mo>(</mo>
<msub>
<mi>B</mi>
<mrow>
<mi>x</mi>
<mi>t</mi>
</mrow>
</msub>
<mi>P</mi>
<mo>(</mo>
<mrow>
<msub>
<mi>w</mi>
<mi>t</mi>
</msub>
<mo>|</mo>
<msub>
<mi>C</mi>
<mi>i</mi>
</msub>
</mrow>
<mo>)</mo>
<mo>+</mo>
<mo>(</mo>
<mrow>
<mn>1</mn>
<mo>-</mo>
<msub>
<mi>B</mi>
<mrow>
<mi>x</mi>
<mi>t</mi>
</mrow>
</msub>
</mrow>
<mo>)</mo>
<mo>(</mo>
<mrow>
<mn>1</mn>
<mo>-</mo>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mrow>
<msub>
<mi>w</mi>
<mi>t</mi>
</msub>
<mo>|</mo>
<msub>
<mi>C</mi>
<mi>i</mi>
</msub>
</mrow>
<mo>)</mo>
</mrow>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
</mrow>
Wherein, | V | the total quantity of representation eigenvalue, BxtIt is wtThe mark occurred in text X, if wtThere is then BxtIt is designated as 1, it is no
Then it is designated as 0, wtRepresent t-th of feature, i.e. t-th vectorial of component, therefore, the P (w in formulat|Ci) calculation formula is as follows：
(2) multinomial model (Multinomial Model) the then occurrence number of statistical nature attribute in the text, calculation formula
For：
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>|</mo>
<msub>
<mi>C</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Pi;</mo>
<mrow>
<mi>t</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>|</mo>
<mi>v</mi>
<mo>|</mo>
</mrow>
</munderover>
<mfrac>
<mrow>
<mi>P</mi>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>w</mi>
<mi>t</mi>
</msub>
<mo>|</mo>
<msub>
<mi>C</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<msub>
<mi>N</mi>
<mrow>
<mi>x</mi>
<mi>t</mi>
</mrow>
</msub>
</msup>
</mrow>
<mrow>
<msub>
<mi>N</mi>
<mrow>
<mi>x</mi>
<mi>t</mi>
</mrow>
</msub>
<mo>!</mo>
</mrow>
</mfrac>
</mrow>
Wherein, NxtRepresent the number of times that feature t occurs in text X；P(wt|Ci) calculation formula it is as follows：
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>w</mi>
<mi>t</mi>
</msub>
<mo>|</mo>
<msub>
<mi>C</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<mn>1</mn>
<mo>+</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>|</mo>
<mi>D</mi>
<mo>|</mo>
</mrow>
</munderover>
<msub>
<mi>N</mi>
<mrow>
<mi>j</mi>
<mi>t</mi>
</mrow>
</msub>
<mi>P</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>C</mi>
<mi>i</mi>
</msub>
<mo>|</mo>
<msub>
<mi>d</mi>
<mi>j</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mo>|</mo>
<mi>V</mi>
<mo>|</mo>
<mo>+</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>s</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>|</mo>
<mi>V</mi>
<mo>|</mo>
</mrow>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>|</mo>
<mi>D</mi>
<mo>|</mo>
</mrow>
</munderover>
<msub>
<mi>N</mi>
<mrow>
<mi>j</mi>
<mi>s</mi>
</mrow>
</msub>
<mi>P</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>C</mi>
<mi>i</mi>
</msub>
<mo>|</mo>
<msub>
<mi>d</mi>
<mi>j</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>;</mo>
</mrow>
In P (wt|Ci) calculation formula in, NjtT is characterized in text djIn the number of times that once occurred, | D | for training text sum, |
V | it is characterized sum, NjsIt is feature s in text djIn occurrence number；The essence of the sorting technique is to the institute in text object
There is the probability that characteristic value is counted and is mapped in each already present classification.
2. archives according to claim 1 are compiled and grind method automatically, it is characterised in that the Compiling of Files resume module being related to
Data message is for the archive information typing set up and archives list, list in volume and the expert's registration card mesh of management module
The basic volume that record is carried out is ground, and being created in archive information typing and management module includes archives list database, list in volume number
According to storehouse, classification of documents database, expert's essential information registered database, civilized thin and 6 data of expert's Item Detail of Specialists
Storehouse；Compiling of Files module is compiled to grind submodule, file font size index volume and grind submodule and expert info and compile by the classification of documents grinds 3 sons
Module is constituted, and the Compiling of Files basic database associated with above-mentioned 6 databases is created in Compiling of Files module；Archives point
Class is compiled and grinds the compiling and researching condition that submodule is inputted according to user, realizes that classification of documents volume is ground, archives title is compiled when grinding and filing automatically
Between compile grind and with tabular form show volume grind result；File font size index volume grinds submodule and grinds demand input bag according to the volume of user
Official document font size, year, shelves number and the combination of retention period compiling and researching condition are included, according to condition filter data after click inquiry, so as to
Automatically generate and show that file font size index volume grinds list；Expert info is compiled and grinds submodule to expert's registration card in file administration
Information carries out volume and grinds statistics, and fuzzy query is carried out according to the compiling and researching condition of expert's name of input, research direction and achievement, realizes
Expert classification volume is ground, expert's research direction volume is ground, expert's paper information is compiled to grind to compile with expert's project information and ground, and volume is ground into result
It is shown in after being collected on screen；Wherein expert classification, which is compiled to grind to refer to compile, grinds all experts for automatically generating a certain research field
Information is compiled and grinds the results list, and its volume, which grinds result, can export to preservation or printout in Excel or Word document.
3. archives according to claim 1 are compiled and grind method automatically, it is characterised in that the classification of documents being related to is compiled and grinds submodule
Compiling of Files comprise the following steps：
(1) create classification volume and grind view, view is using list in volume or archives list as main table, and associative classification information table obtains classification
Name of the information, list in volume is stored in from archives list information in different tables of data, is carrying out needing to collect when classification volume is ground
Two parts information, carries out unified query and retrieves；
(2) data access layer identification code, retrieval needs to compile the archives letter ground during data access layer is the view provided from step (1)
Breath, performs function using querying condition as parameter, retrieves qualified archive information, and classification volume grinds the extraction classification of documents, archives
Title, time of filing, number, number of pages and shelves information, and result is ground according to specific name, distribution caption and time of filing to volume
It is ranked up；
(3) classification volume grinds application layer realization, classification compiling and researching condition is set first, by the fuzzy search of classification of documents title, archives mark
Topic fuzzy search and time of filing searching classification, which are compiled, to be ground；
(4) volume is ground into result to export in EXCEL or Word document, facilitates user's preservation to check and print and bind into book form.
4. archives according to claim 1 are compiled and grind method automatically, it is characterised in that the file font size index volume being related to grinds son
The Compiling of Files of module comprises the following steps：
(1) establishment file font size indexed view, file font size indexed view is carried using list in volume as main table associative classification information table
Take official document font size, year, file sequence number, shelves number, page number, number of pages and retention period information；
(2) data access layer identification code, data access layer is that information is extracted from file font size indexed view, and according to official document word
Number, file sequence number and year sequence, function is using querying condition as parameter, by application layer dynamic construction；
(3) application layer main code, default compiling and researching condition, including official document font size, year, shelves number and retention period, according to
Volume grinds needs and is combined input, clicks on after inquiry according to condition filter data；
(4) volume is ground into result to export in EXCEL or Word document, facilitates user's preservation to check and print and bind into book form.
5. archives according to claim 1 are compiled and grind method automatically, it is characterised in that the expert info being related to is compiled and grinds submodule
Compiling of Files comprise the following steps：
(1) create expert info volume and grind view, view is using expert's essential information registration form as main table, the literary detail list of association Specialists
Paper information is extracted, association expert item details extract project and winning information, the registration of expert's essential information and expert's achievement
Information is stored in different tables of data, is needed to collect each several part information when row information is compiled and ground entering, is carried out unified query；
(2) data access layer identification code, data access layer is to obtain to need to compile the expert info ground from the view in step (1),
Function inquires about qualified expert's archive information using querying condition as parameter, and volume is ground extraction expert name, expert's classification, ground
Study carefully direction, paper information and project information；
(3) expert info, which is compiled, grinds application layer realization, sets compiling and researching condition, carries out the fuzzy search of expert's title, expert's research direction
Fuzzy search, thesis topic summarizes fuzzy search with paper, and paper publishing time retrieval, project name fuzzy search, project is general
State fuzzy search and the retrieval of project beginning and ending time and the prize-winning situation retrieval of project；Expert info is compiled to grind grinds bar according to the volume of input
Part, realizes that expert classification volume is ground, expert's research direction volume is ground, expert's paper information is compiled to grind to compile with expert's project information and ground；
(4) volume is ground into result to export in Excel or Word document, facilitates user's preservation to check and print and bind into book form.
6. archives according to claim 1 are compiled and grind method automatically, it is characterised in that the MIS of Department Files being related to
Performing flow is：
(1) a browser is opened in client, the station address of input system sends clothes to Web server in address field
Business request, when showing the login page of system on desktop, is transmitted after login page fills in user name, password and identifying code
To Web server, Web server passes the homepage of MIS of Department Files with http protocol after being verified to user identity
Client is given, client browser receives the homepage file transmitted, and it is shown on screen；
(2) typing of archives essential information, according to system be shown in the archives list, list in volume and expert's registration card of homepage to
Classification, shelves number, archives title, year and all kinds of essential informations of expert belonging to the unified form difference typing gone out and addition archives；
System performs corresponding extension application in the Business Logic of Web server and is attached with database server, passes through
The above-mentioned all kinds of essential informations of user's typing or addition are arrived the corresponding number being connected with Web server by SQL modes in storage
Before storehouse, catalogue and archives title are carried out automatically classifying and filing by system, then original text is mounted；Original text can be electronics
Scanned copy or electronic edition original paper；
(3) when needing to carry out Compiling of Files to certain class archive information, the entry for the Compiling of Files that system is selected according to user enters
To corresponding Compiling of Files interface, compiling and researching condition is inputted in the interface and clicks on inquiry button, associated databases are believed
Breath retrieval and inquiry；
(4) volume inputted according to step (3) grinds parameter, performed with Web server link after by SQL statement to corresponding
Database server propose data processing request, i.e., to Compiling of Files basic database and be associated other database informations
Retrieved and inquiry operation, and the data item that meets compiling and researching condition will be retrieved and counted, analyzed and collected, generate archives
Volume is ground into fruit；
(5) result of the Compiling of Files of generation is submitted to Web server by database server, then is sent to by Web server
Client is simultaneously shown on screen；
(6) volume is ground into fruit and exports to preservation or printout in Word document or Excel forms.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410714594.2A CN104361111B (en) | 2014-11-28 | 2014-11-28 | A kind of archives are compiled and grind method automatically |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410714594.2A CN104361111B (en) | 2014-11-28 | 2014-11-28 | A kind of archives are compiled and grind method automatically |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104361111A CN104361111A (en) | 2015-02-18 |
| CN104361111B true CN104361111B (en) | 2017-10-27 |
Family
ID=52528371
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410714594.2A Active CN104361111B (en) | 2014-11-28 | 2014-11-28 | A kind of archives are compiled and grind method automatically |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104361111B (en) |
Families Citing this family (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105303321A (en) * | 2015-11-04 | 2016-02-03 | 广州赛莱拉干细胞科技股份有限公司 | Archive management method and apparatus |
| CN105808770A (en) * | 2016-03-22 | 2016-07-27 | 北京北方微电子基地设备工艺研究中心有限责任公司 | File management method and device |
| CN106021355B (en) * | 2016-05-10 | 2020-07-28 | 重庆大学 | Statistical method and custom rule establishing method, device and system among multiple tables |
| CN106227748A (en) * | 2016-07-14 | 2016-12-14 | 上海超橙科技有限公司 | A kind of information generating method and equipment |
| CN106227749A (en) * | 2016-07-14 | 2016-12-14 | 上海超橙科技有限公司 | A kind of information-pushing method and equipment |
| CN106776695B (en) * | 2016-11-11 | 2020-12-04 | 上海信联信息发展股份有限公司 | Method for automatically identifying value of document and file |
| CN107463651A (en) * | 2017-07-27 | 2017-12-12 | 合肥泓泉档案信息科技有限公司 | A kind of electronic record is filed management method |
| CN107491498A (en) * | 2017-07-27 | 2017-12-19 | 合肥泓泉档案信息科技有限公司 | A kind of automatic adjusting method of dossier table |
| CN109684608A (en) * | 2017-10-19 | 2019-04-26 | 航天信息股份有限公司 | It is a kind of that the method and system of generation EXCEL document are passed through based on database |
| CN107894999A (en) * | 2017-10-27 | 2018-04-10 | 成都准星云学科技有限公司 | Towards the topic type automatic classification method and system based on thinking of solving a problem of elementary mathematics |
| CN107943957A (en) * | 2017-11-27 | 2018-04-20 | 广西简约科技有限公司 | A kind of software design approach for collecting meeting summary |
| CN108763467B (en) * | 2018-05-29 | 2023-07-11 | 甘肃集优品网络科技有限公司 | Electronic file intelligent processing management system suitable for archives trade |
| CN109189730A (en) * | 2018-09-21 | 2019-01-11 | 郑州云海信息技术有限公司 | A kind of archives visual management method, system, device and readable storage medium storing program for executing |
| CN109766439A (en) * | 2018-12-15 | 2019-05-17 | 内蒙航天动力机械测试所 | The unlimited tree-shaped class definition and assigning method of statistical query software |
| CN111597150B (en) * | 2020-05-09 | 2023-09-12 | 云南驰宏锌锗股份有限公司 | Automatic change and file arrangement information system |
| CN111858499A (en) * | 2020-08-03 | 2020-10-30 | 王洋 | File identification method, system and device based on black and white list |
| CN112463896B (en) * | 2020-12-08 | 2024-02-23 | 常兰会 | Archive catalogue data processing method, archive catalogue data processing device, computing equipment and storage medium |
| CN112861473B (en) * | 2021-03-12 | 2024-02-02 | 国网浙江省电力有限公司物资分公司 | Directory examination result summarizing system and method based on openpyl |
| CN113204610A (en) * | 2021-05-06 | 2021-08-03 | 广东博维创远科技有限公司 | Automatic cataloguing method based on criminal case electronic file and computer readable storage device |
| CN113407645A (en) * | 2021-05-19 | 2021-09-17 | 福建福清核电有限公司 | Intelligent sound image archive compiling and researching method based on knowledge graph |
| CN113220842B (en) * | 2021-05-20 | 2022-04-19 | 广州中海云科技有限公司 | Processing method, device and equipment for maritime affair administration punishment cutting template |
| CN113590903B (en) * | 2021-09-27 | 2022-01-25 | 广东电网有限责任公司 | Management method and device of information data |
| CN114947402A (en) * | 2022-06-20 | 2022-08-30 | 国网山东省电力公司冠县供电公司 | Archives screening classification device |
| CN115329086B (en) * | 2022-08-29 | 2024-04-16 | 中铁四局集团电气化工程有限公司 | Track traffic document retrieval system and method based on classification coding |
| CN116501862B (en) * | 2023-06-25 | 2023-09-12 | 桂林电子科技大学 | Automatic text extraction system based on dynamic distributed collection |
| CN116595238B (en) * | 2023-07-17 | 2023-09-19 | 三土电子有限公司 | User archive data analysis processing method based on RFID technology |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102368273A (en) * | 2011-11-29 | 2012-03-07 | 神华集团有限责任公司 | Archive management system and method |
| CN103745302A (en) * | 2013-12-19 | 2014-04-23 | 镇江锐捷信息科技有限公司 | Digitalized archival data management system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090132496A1 (en) * | 2007-11-16 | 2009-05-21 | Chen-Kun Chen | System And Method For Technique Document Analysis, And Patent Analysis System |
-
2014
- 2014-11-28 CN CN201410714594.2A patent/CN104361111B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102368273A (en) * | 2011-11-29 | 2012-03-07 | 神华集团有限责任公司 | Archive management system and method |
| CN103745302A (en) * | 2013-12-19 | 2014-04-23 | 镇江锐捷信息科技有限公司 | Digitalized archival data management system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104361111A (en) | 2015-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104361111B (en) | A kind of archives are compiled and grind method automatically | |
| CN104462306B (en) | A kind of archives compile grinding device automatically | |
| CN109992645B (en) | Data management system and method based on text data | |
| CN103914478B (en) | Webpage training method and system, webpage Forecasting Methodology and system | |
| CN101593200B (en) | Method for classifying Chinese webpages based on keyword frequency analysis | |
| US9317613B2 (en) | Large scale entity-specific resource classification | |
| CN100440224C (en) | Automatization processing method of rating of merit of search engine | |
| CN103226578B (en) | Towards the website identification of medical domain and the method for webpage disaggregated classification | |
| US20070198459A1 (en) | System and method for online information analysis | |
| CN101609450A (en) | Web page classification method based on training set | |
| CN112632405B (en) | Recommendation method, recommendation device, recommendation equipment and storage medium | |
| CN106383887A (en) | Environment-friendly news data acquisition and recommendation display method and system | |
| CN104077407B (en) | A kind of intelligent data search system and method | |
| CN101794311A (en) | Fuzzy data mining based automatic classification method of Chinese web pages | |
| CN110532309B (en) | Generation method of college library user portrait system | |
| Irudeen et al. | Big data solution for Sri Lankan development: A case study from travel and tourism | |
| CN111192176B (en) | Online data acquisition method and device supporting informatization assessment of education | |
| KR100974064B1 (en) | System for providing information adapted to users and method thereof | |
| Kim et al. | Event diffusion patterns in social media | |
| KR101801257B1 (en) | Text-Mining Application Technique for Productive Construction Document Management | |
| CN103810162A (en) | Method and system for recommending network information | |
| CN104834739B (en) | Internet information storage system | |
| Balasubramaniam et al. | Identifying covid-19 misinformation tweets and learning their spatio-temporal topic dynamics using nonnegative coupled matrix tensor factorization | |
| CN116010552A (en) | Engineering cost data analysis system and method based on keyword word library | |
| KR20170043365A (en) | Important precedents extraction and sorting method using Big Data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |