CN107577922A - A kind of corn lncRNA sifting sort methods based on arm processor - Google Patents
A kind of corn lncRNA sifting sort methods based on arm processor Download PDFInfo
- Publication number
- CN107577922A CN107577922A CN201710848923.6A CN201710848923A CN107577922A CN 107577922 A CN107577922 A CN 107577922A CN 201710848923 A CN201710848923 A CN 201710848923A CN 107577922 A CN107577922 A CN 107577922A
- Authority
- CN
- China
- Prior art keywords
- lncrna
- corn
- sequence
- prediction
- arm processor
- 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
Landscapes
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
- Breeding Of Plants And Reproduction By Means Of Culturing (AREA)
Abstract
The invention discloses a kind of corn lncRNA sifting sort methods based on arm processor, comprise the following steps:Step 1: corn full length cDNA sequence is inputted into CPLD;Step 2: screening length is more than 200 cDNA sequence, ORFs prediction and threshold value are carried out to cDNA, determine that length is less than the cDNA of threshold value, its make with known albumen it is homologous compare, not analogous parts are retained, secondary structure prediction is made to existing plant lncRNA, second structure characteristic is extracted, while makees secondary structure prediction to residue sequence, retains the part for meeting architectural feature, cDNA sequence to be screened is compared with tiny RNA sequence, excludes tiny RNA precursor;Step 3: collect the corn lncRNA sequences of known function, and carry out structure prediction, architectural feature extraction is carried out to function lncRNA and is screened, build known function corn lncRNA structure characteristic collections, the corn lncRNA that step 2 is filtered out carries out secondary structure extraction, structure characteristic collection is compared, lncRNA functions are confirmed to classify.
Description
Technical field
The present invention relates to bioinformatics research field, and in particular to a kind of corn lncRNA sieves based on arm processor
Select sorting technique.
Background technology
RNA secondary structure just refers to because base pair complementarity, RNA itself fold the stem-loop structure formed.Because base is mutual
Recruit pair, although RNA is that the single-stranded subregion that itself can but be folded in forms double-spiral structure so that RNA three dimensions knot
Structure becomes complicated.Except typical case's pairing (Waston-Crick is matched, A=U, G ≡ C) outside, also G=U (wobble base
Pairs swings pairing) this exception.G ≡ C stability highests, G=U stability are minimum.Stem is base in secondary structure
The continuous place for forming pairing, the double helix region formed according to above-mentioned A=U, G ≡ C, G=U pairing rules.Part does not have
The single stranded portion for forming pairing is ring.
When studying relevant RNA the problem of, we can obtain the partial information about RNA by RNA primary sequences,
And the research to RNA secondary structures can further explore its biological function having.RNA secondary structure predictions are present lifes
The focal issue of thing informatics, is developed into today, has had many algorithms to solve this problem.Summarize conventional two level knot
Structure prediction algorithm has following 4 class:Dynamic programming algorithm, comparative sequence analysis method method, combinatorial optimization algorithm, heuritic approach.At present
Realize that RNA secondary structure predictions take time and effort using BIOLOGICAL TEST METHODS, and to be confined to sequence whole for the prediction based on computational methods
The accuracy of body structure, ignore the conservative functional unit of partial structurtes, be not suitable for the forecast demand of long sequence.To meet
The demand of lncRNA secondary structure predictions, so proposing lncRNA segmentation secondary structure prediction new methods.
Sequence alignment algorithms mainly have three, enumerative technique, dynamic programming algorithm, blast algorithms.Enumerative technique is to be easiest to think
Arrive and an intelligible method, but the method does not have any optimization, time complexity is higher, can not be counted within the reasonable time
Calculate.So enumerative technique can only be theoretical feasible, can not practical application in longer sequence.Dynamic programming algorithm is asked in solution decision-making
A kind of method of optimization during topic.Dynamic Programming is when for optimization problem, has a variety of feasible solutions, all feasible solutions
There is the value of oneself, the value for finding optimal solution is exactly the effect of Dynamic Programming.Its thought is to split big problem, is solved respectively
Afterwards, then by solution merge, the non-optimal solution that majority is not required to calculate is removed in solution procedure, saves many times.Consider based on dynamic
The comparison of state planning algorithm, its time complexity are proportional to two sequences, i.e. amount of calculation is equivalent to calculating with two sequences length
The two-dimensional matrix of foundation.If can reduce the amount of calculation of part in two-dimensional matrix can then lift calculating speed, blast algorithms
Calculated as far as possible around optimal comparison path, reduce many unnecessary computings.
The content of the invention
The present invention has designed and developed a kind of corn lncRNA sifting sort methods based on arm processor, hair of the invention
Improving eyesight is to reduce operation time, the problem of improving breakneck acceleration and screen accuracy.
Technical scheme provided by the invention is:
A kind of corn lncRNA sifting sort methods based on arm processor, comprise the following steps:
Step 1: corn full length cDNA sequence will be inputted into CPLD, it is by encoding contingency table by cDNA sequence to encode
Form be stored in SRAM;
Step 2: screening length is more than 200 cDNA sequence, ORFs prediction and threshold value are carried out to cDNA,
Determine that length is less than the cDNA of the threshold value, its make with known albumen it is homologous compare, not analogous parts are retained, to existing
Plant lncRNA makees secondary structure prediction, extracts second structure characteristic, while makees secondary structure prediction to residue sequence, retains full
The part of sufficient architectural feature, cDNA sequence to be screened is compared with tiny RNA sequence, excludes tiny RNA precursor;
Step 3: collecting the corn lncRNA sequences of known function, and structure prediction is carried out, function lncRNA is carried out
Architectural feature is extracted and screened, and builds known function corn lncRNA structure characteristic collections, the corn that the step 2 is filtered out
LncRNA carries out secondary structure extraction, compares the structure characteristic collection, and the lncRNA functions are confirmed to classify.
Preferably, in the step 2, reading frame prediction is comprised the following steps:The existing lncRNA of prediction with
The ORFs of coding RNA, coding and non-coding are distinguished according to ORFs length, setting ORFs length is reasonable
Threshold value simultaneously filters out the sequence less than this threshold value.
Preferably, in the step 2, the tiny RNA includes miRNA, shRNA, siRNA.
Preferably, in the step 2, the sequence to be screened passes through blast algorithms with the tiny RNA sequence
It is compared.
Preferably, in the step 2, the lncRNA secondary structure predictions are comprised the following steps:
LncRNA segmentations based on transcription time sequencing, the divided stages that each segment structure is formed, for the spiral being likely to form
Area is given a mark, and the helical region for judging most urgently to be formed according to score value difference, is obtained by dynamic programming algorithm using helical region to be single
The secondary structure of position, determines that optimum structure forms the lncRNA secondary structures of the prediction.
Preferably, the lncRNA section lengths are 160nt, every section with front and rear segmentation overlay 80nt.
Preferably, in the step 3, corn lncRNA sequences classification is also comprised the following steps:According to alkali
Base normalization to base pairing and does not form the normalization of each base quantity of pairing and calculated to characteristics algorithm;Foundation
Second structure characteristic algorithm, calculate normalizing, being average for hairpin loop in normalization of the minimum MFE to sequence, secondary structure
The characteristic features such as base pairing of the hairpin structure in the stem area normalization longer than, secondary structure middle ring;It is special to the typical case of extraction
Sign carries out quantitative consistency analysis.
Preferably, in the step 3, corn lncRNA sequences classification is also comprised the following steps:Collect
Know the characteristic feature that a certain function is extracted and the corn lncRNA for determining not having this function;Using SVM modeling methods, to mould
Type training is simultaneously assessed;The feature that ranking is located at first 3, the feature set as a certain function are assessed in screening.
Present invention beneficial effect possessed compared with prior art:
1st, high-performance, low-power consumption, high performance-price ratio, high code density.Arm processor is directed to Embedded Application at present, full
On the premise of sufficient performance, lowest power consumption also can reach.It can take into account performance, power consumption, code density and the several sides of price
Face;
2nd, abundant chip.At present, the various processors based on ARM kernels have hundreds of.User can be according to respective
Application demand, binding ability, function etc. are considered, and are chosen suitable chip and are designed the application system of oneself;
Corn lncRNA sifting sort is realized using arm processor, not only can be in order to batch execution data, while can
To take into account its higher processing speed, and higher accuracy rate, and can be in order to being cured to high flux lncRNA numbers
It is not only convenient and swift according in acquisition instrument, standardization, this screening process of modularization, and manual intervention can be reduced.
Brief description of the drawings
Fig. 1 is the flow chart of the corn lncRNA sifting sort methods of the present invention based on arm processor.
Fig. 2 is the structural representation of the corn lncRNA sifting sort methods of the present invention based on arm processor.
Embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings, to make those skilled in the art with reference to specification text
Word can be implemented according to this.
As shown in Figure 1 and Figure 2, existing lncRNA sifting sort method, it is high to be first according to traditional design rule collection
The lncRNA data of flux, then carry out sifting sort etc. to it.Collection lncRNA needs to be carried out with Bioexperiment instrument, adopts
With current method, it is not easy to be combined with high flux data acquisition instrument etc., and cost of labor is high, speed is slow, and efficiency is low, because
This, is badly in need of lncRNA sorting techniques being cured in high flux data acquisition instrument, optimizes lncRNA sifting sorts;Therefore, originally
Invention provides a kind of corn lncRNA sifting sort methods based on arm processor, the hardware device bag required for this method
Include processor, internal memory, mainboard;The present invention comprises the following steps:
1st, cDNA sequence is inputted into CPLD;
Wherein, CPLD (Complex Programmable Logic Device) is CPLD, from
PAL and GAL device developments, comparatively scale is big, complicated, belongs to large scale integrated circuit scope, is a kind of use
Family according to respective the need and voluntarily digital integrated electronic circuit of constitutive logic function.CPLD basic design methods are soft by Integrated Development
Part platform, the methods of with schematic diagram, hardware description language, corresponding file destination is generated, passing through download cable, (" in system " is compiled
Journey) code is sent in objective chip to realize design.PAL refers to programmable logic array, is the end of the seventies by MMI companies
Take the lead in release a kind of low-density, One Time Programmable logical device.GAL is GAL, is developed from PAL, because
Make it that the programming of the device is very convenient to employ EECMOS techniques;
2nd, cDNA sequence is stored in SRAM by CPLD in an encoded form by encoding contingency table.We are special according to sequence
Architectural feature of seeking peace screens corn lncRNA jointly;
Wherein, SRAM is English Static RAM abbreviation, and it is a kind of internal memory with static access facility, it is not necessary to
Refresh circuit can preserve the data of its storage inside.We screen corn jointly according to sequence signature and architectural feature
lncRNA。
3rd, corn lncRNA screenings are carried out from ARM microprocessor:
1) by digital information corresponding to corn full length cDNA sequence, it is loaded into screening model;
2) cDNA sequence that length is more than 200 is screened;
3) ORFs prediction is carried out to cDNA;The existing lncRNA of prediction and coding RNA ORFs, foundation are opened
Put reading frame length and distinguish coding and non-coding, set ORFs length reasonable threshold value, filter out the sequence less than this threshold value
Row;
4) cDNA and known albumen are made into homologous comparison, retains not homologous part;
5) make secondary structure prediction to existing plant lncRNA, extract second structure characteristic;
6) make secondary structure prediction to residue sequence, retain the part for meeting architectural feature;
7) it is the Mirnas of plant (Microrna), shRNA (short hairpin RNA), siRNA (siRNA) of collection etc. is small
RNA sequence is stored in SRAM, and using blast algorithms, sequence to be screened is compared with tiny RNA sequence, is deleted wherein
Possible tiny RNA precursor;
Wherein, transcript length:Sequence length 200 is a conclusive spy for determining whether long-chain non-coding RNA
Sign;ORF length:ORF (ORFs, Open Reading Frame) is to have coding in DNA since initiation codon
The potential of protein, one section of base sequence without terminator codon termination.Because the initiation site of codon read-write is different, its
Sequence has six kinds of possible readings and translation (having two chains, every chain has three kinds);With the homology of known albumen:Only
ORF feature might not can characterize the ability of protein coding, it is also necessary to include the feature of the homology with known albumen;
4th, corn lncRNA classification is carried out from ARM microprocessor:
1) the corn lncRNA sequences of known function are collected, and carry out structure prediction;
2) architectural feature extraction is carried out to function lncRNA and is screened, the result of screening is output to LCD liquid crystal displays
Device;
3) modeled using SVM, build known function corn lncRNA structure characteristic collections, negative sample does not have this work(for determination
The lncRNA of energy.Simultaneously by cross-validation method, ranking is assessed to structure characteristic collection, using ranking first three as final feature set
Element;
4) secondary structure extraction is carried out to the corn lncRNA of unknown function;
5) the function corn lncRNA feature sets established are compared, see its matching degree, matching degree height can then be judged to have in advance
This function, functional character identification is carried out to unknown function lncRNA with this;
6) unknown lncRNA functions confirm to realize classification;
Wherein, base-pair feature:Normalization including base pairing and do not formed pairing each base quantity normalization
Calculate, it can intuitively determine architectural feature, be one of simplest architectural feature;Second structure characteristic:Including stem area, interior
The architectural feature of the elementary cells such as ring, hairpin loop, multi-branched ring, and normalization of the minimum MFE to sequence, secondary structure
The allusion quotations such as the normalizing of middle hairpin loop, average base pairing of the hairpin structure in stem the area normalization longer than, secondary structure middle ring
Type feature.Extraction to characteristic feature, the difference in structure between each function lncRNA is can determine, preferably lncRNA is entered
Row function classification.
As shown in Figure 1 and Figure 2, the invention provides a kind of corn lncRNA sifting sort methods based on arm processor,
Comprise the following steps:
Step 1: the second structure characteristic of long-chain non-coding RNA in corn is excavated, then by corn full length cDNA sequence
CPLD is inputted, possible long-chain non-coding is found as criterion of identification by the use of the sequence signature and architectural feature of long-chain non-coding RNA
RNA;
Step 2: cDNA sequence is stored in SRAM in the form of coded sequence, then with ARM processors, carry out
LncRNA is screened and classification;
Step 3: carrying out first time screening according to sequence length, the cDNA sequence that length is more than 200 is obtained;
Step 4: carrying out the statistical analysis of ORF sizes to existing long-chain non-coding RNA afterwards, rational threshold value is set,
Programmed screening is carried out according to this feature, obtains the sequence less than this threshold value;
Step 5: the data of previous step identification are done into Homology search with known albumen, third time screening is carried out, is obtained non-same
Source sequence;
Step 6: the tiny RNA sequence of collection is stored in SRAM, using blast algorithms, the local for comparing and needing is established
Database, using long-chain non-coding RNA to be screened as database, and tiny RNA is aligned sequences, is removed wherein possible small
RNA precursors;
Step 7: carrying out structural analysis to known function corn lncRNA, functional structure feature is established using svm
Collection, the unknown function lncRNA filtered out is subjected to feature extraction;
Step 8: comparing known features collection, the most at last high lncRNA of matching degree is determined as thering is the function, finally will knot
Fruit is output to LCD liquid crystal displays.
Although embodiment of the present invention is disclosed as above, it is not restricted in specification and embodiment listed
With it can be applied to various suitable the field of the invention completely, can be easily for those skilled in the art
Other modification is realized, therefore under the universal limited without departing substantially from claim and equivalency range, it is of the invention and unlimited
In specific details and shown here as the legend with description.
Claims (8)
- A kind of 1. corn lncRNA sifting sort methods based on arm processor, it is characterised in that comprise the following steps:Step 1: corn full length cDNA sequence is inputted into CPLD, it is by encoding contingency table by cDNA sequence in an encoded form It is stored in SRAM;Step 2: screening length is more than 200 cDNA sequence, ORFs prediction and threshold value are carried out to cDNA, it is determined that Length is less than the cDNA of the threshold value, its make with known albumen it is homologous compare, not analogous parts are retained, to existing plant LncRNA makees secondary structure prediction, extracts second structure characteristic, while makees secondary structure prediction to residue sequence, retains and meets knot The part of structure feature, cDNA sequence to be screened is compared with tiny RNA sequence, excludes tiny RNA precursor;Step 3: collecting the corn lncRNA sequences of known function, and structure prediction is carried out, it is special to carry out structure to function lncRNA Sign is extracted and screened, and builds known function corn lncRNA structure characteristic collections, the corn lncRNA that the step 2 is filtered out Secondary structure extraction is carried out, compares the structure characteristic collection, the lncRNA functions are confirmed to classify.
- 2. the corn lncRNA sifting sort methods based on arm processor as claimed in claim 1, it is characterised in that in institute State in step 2, reading frame prediction is comprised the following steps:The existing lncRNA of prediction and coding RNA ORFs, Coding and non-coding are distinguished according to ORFs length, ORFs length reasonable threshold value is set and filters out less than this threshold The sequence of value.
- 3. the corn lncRNA sifting sort methods based on arm processor as claimed in claim 1 or 2, it is characterised in that In the step 2, the tiny RNA includes miRNA, shRNA, siRNA.
- 4. the corn lncRNA sifting sort methods based on arm processor as claimed in claim 3, it is characterised in that in institute State in step 2, the sequence to be screened is compared with the tiny RNA sequence by blast algorithms.
- 5. the corn lncRNA sifting sort methods based on arm processor as claimed in claim 4, it is characterised in that in institute State in step 2, the lncRNA secondary structure predictions are comprised the following steps:LncRNA segmentations based on transcription time sequencing, the divided stages that each segment structure is formed, the helical region to be likely to form is entered Row marking, the helical region for judging most urgently to be formed according to score value difference, is obtained in units of helical region by dynamic programming algorithm Secondary structure, determine that optimum structure forms the lncRNA secondary structures of the prediction.
- 6. the corn lncRNA sifting sort methods based on arm processor as claimed in claim 5, it is characterised in that described LncRNA section lengths are 160nt, every section with front and rear segmentation overlay 80nt.
- 7. the corn lncRNA sifting sort methods based on arm processor as claimed in claim 6, it is characterised in that in institute State in step 3, corn lncRNA sequences classification is also comprised the following steps:According to base-pair characteristics algorithm, base is matched somebody with somebody To normalization and do not form the normalization of each base quantity of pairing and calculate;Foundation second structure characteristic algorithm, calculate The normalizing of hairpin loop, average base of the hairpin structure in stem area in normalization of the minimum MFE to sequence, secondary structure With characteristic features such as contrast, the normalization of secondary structure middle ring length;Quantitative uniformity point is carried out to the characteristic feature of extraction Analysis.
- 8. the corn lncRNA sifting sort methods based on arm processor as claimed in claim 7, it is characterised in that in institute State in step 3, corn lncRNA sequences classification is also comprised the following steps:The allusion quotation that a certain function known to collection is extracted Type feature and the corn lncRNA for determining not having this function;Using SVM modeling methods, to model training and assess;Screening is assessed Ranking is located at the feature of first 3, the feature set as a certain function.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710848923.6A CN107577922B (en) | 2017-09-20 | 2017-09-20 | ARM processor-based corn lncRNA screening and classifying method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710848923.6A CN107577922B (en) | 2017-09-20 | 2017-09-20 | ARM processor-based corn lncRNA screening and classifying method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107577922A true CN107577922A (en) | 2018-01-12 |
CN107577922B CN107577922B (en) | 2020-07-03 |
Family
ID=61033121
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710848923.6A Expired - Fee Related CN107577922B (en) | 2017-09-20 | 2017-09-20 | ARM processor-based corn lncRNA screening and classifying method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107577922B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108614955A (en) * | 2018-05-04 | 2018-10-02 | 吉林大学 | One kind is formed based on sequence, the lncRNA identification methods of structural information and physicochemical characteristics |
CN108846262A (en) * | 2018-05-31 | 2018-11-20 | 广西大学 | The method that RNA secondary structure distance based on DFT calculates phylogenetic tree construction |
CN111354420A (en) * | 2020-03-08 | 2020-06-30 | 吉林大学 | siRNA research and development method for COVID-19 virus drug therapy |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102880811A (en) * | 2012-10-24 | 2013-01-16 | 吉林大学 | Method for predicting secondary structure of ribonucleic acid (RNA) sequence based on complex programmable logic device (CPLD) base fragment encoding and ant colony algorithm |
CN103020489A (en) * | 2013-01-04 | 2013-04-03 | 吉林大学 | Novel method for forecasting siRNA interference efficiency based on ARM (Advanced RISC Machines) microprocessor |
CN103593587A (en) * | 2013-11-20 | 2014-02-19 | 吉林大学 | Component-based identification method and device of long-chain RNA secondary structures with pseudo knots |
CN105385706A (en) * | 2015-12-11 | 2016-03-09 | 浙江海洋学院 | Eukaryotic expression method of sea cucumber cathepsin |
-
2017
- 2017-09-20 CN CN201710848923.6A patent/CN107577922B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102880811A (en) * | 2012-10-24 | 2013-01-16 | 吉林大学 | Method for predicting secondary structure of ribonucleic acid (RNA) sequence based on complex programmable logic device (CPLD) base fragment encoding and ant colony algorithm |
CN103020489A (en) * | 2013-01-04 | 2013-04-03 | 吉林大学 | Novel method for forecasting siRNA interference efficiency based on ARM (Advanced RISC Machines) microprocessor |
CN103593587A (en) * | 2013-11-20 | 2014-02-19 | 吉林大学 | Component-based identification method and device of long-chain RNA secondary structures with pseudo knots |
CN105385706A (en) * | 2015-12-11 | 2016-03-09 | 浙江海洋学院 | Eukaryotic expression method of sea cucumber cathepsin |
Non-Patent Citations (3)
Title |
---|
SUSAN BOERNER等: "Computational Identification and Functional Predictions of Long Noncoding RNA in Zea mays", 《PLOS ONE》 * |
党合萱: "基于多特征的长非编码RNA识别方法", 《中国优秀硕士学位论文全文数据库(电子期刊)基础科学辑》 * |
王金华等: "基于最小自由能和协变信息预测带伪结RNA二级结构的迭代化方法", 《遗传》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108614955A (en) * | 2018-05-04 | 2018-10-02 | 吉林大学 | One kind is formed based on sequence, the lncRNA identification methods of structural information and physicochemical characteristics |
CN108846262A (en) * | 2018-05-31 | 2018-11-20 | 广西大学 | The method that RNA secondary structure distance based on DFT calculates phylogenetic tree construction |
CN111354420A (en) * | 2020-03-08 | 2020-06-30 | 吉林大学 | siRNA research and development method for COVID-19 virus drug therapy |
Also Published As
Publication number | Publication date |
---|---|
CN107577922B (en) | 2020-07-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP3864577B1 (en) | Systems and methods for image preprocessing | |
Iftikhar et al. | An evolution based hybrid approach for heart diseases classification and associated risk factors identification | |
CN111126386B (en) | Sequence domain adaptation method based on countermeasure learning in scene text recognition | |
Ertosun et al. | Automated grading of gliomas using deep learning in digital pathology images: a modular approach with ensemble of convolutional neural networks | |
CN105701513B (en) | The method of rapidly extracting palmmprint area-of-interest | |
CN109670177A (en) | One kind realizing the semantic normalized control method of medicine and control device based on LSTM | |
CN109558487A (en) | Document Classification Method based on the more attention networks of hierarchy | |
CN109740154A (en) | A kind of online comment fine granularity sentiment analysis method based on multi-task learning | |
CN108388651A (en) | A kind of file classification method based on the kernel of graph and convolutional neural networks | |
CN108052911A (en) | Multi-modal remote sensing image high-level characteristic integrated classification method based on deep learning | |
CN107683469A (en) | A kind of product classification method and device based on deep learning | |
CN108804677A (en) | In conjunction with the deep learning question classification method and system of multi-layer attention mechanism | |
CN109471944A (en) | Training method, device and the readable storage medium storing program for executing of textual classification model | |
Aslan et al. | Multi-classification deep CNN model for diagnosing COVID-19 using iterative neighborhood component analysis and iterative ReliefF feature selection techniques with X-ray images | |
CN107577922A (en) | A kind of corn lncRNA sifting sort methods based on arm processor | |
CN108280236A (en) | A kind of random forest visualization data analysing method based on LargeVis | |
WO2023124342A1 (en) | Low-cost automatic neural architecture search method for image classification | |
CN108765374A (en) | A kind of method of abnormal core region screening in cervical smear image | |
CN108920446A (en) | A kind of processing method of Engineering document | |
CN115952292B (en) | Multi-label classification method, apparatus and computer readable medium | |
CN115909011A (en) | Astronomical image automatic classification method based on improved SE-inclusion-v 3 network model | |
CN115146062A (en) | Intelligent event analysis method and system fusing expert recommendation and text clustering | |
CN113535960A (en) | Text classification method, device and equipment | |
CN115600602B (en) | Method, system and terminal device for extracting key elements of long text | |
CN115062602B (en) | Sample construction method and device for contrast learning and computer equipment |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200703 Termination date: 20210920 |