CN106295887A - Lasting seed bank Forecasting Methodology based on random forest - Google Patents
Lasting seed bank Forecasting Methodology based on random forest Download PDFInfo
- Publication number
- CN106295887A CN106295887A CN201610659724.6A CN201610659724A CN106295887A CN 106295887 A CN106295887 A CN 106295887A CN 201610659724 A CN201610659724 A CN 201610659724A CN 106295887 A CN106295887 A CN 106295887A
- Authority
- CN
- China
- Prior art keywords
- seed
- lasting
- bank
- seed bank
- data base
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of lasting seed bank Forecasting Methodology based on random forest.The technical scheme is that collection or measure plant seed character, building seed properties data base；Collect or measure plant lasting seed bank attribute, building lasting seed bank data base；Seed properties data base is associated with lasting seed bank data base；Selected part data from data base, utilize random forests algorithm to build training set；Based on described training set, set up forecast model；From seed properties data base, choose the data not being associated with lasting seed bank data, set up forecast set；Utilize described forecast model, it was predicted that collection is predicted；According to predicting the outcome, it may be judged whether whether corresponding plant exists lasting seed bank.The method can be used for predicting whether plant species exists lasting seed bank, saving manpower, time simple and easy to do compared with experimental technique.
Description
Technical field
The present invention relates to botany and recover field of ecology, particularly to one based on random forests algorithm, for pre-
Whether measuring plants kind exists the method for lasting seed bank.It is to understand plant and environmental interaction, promotes revegetation and species
The important means of Protection of Diversity.
Background technology
Persistently seed bank refers to exist more than 1 year in soil, still has the seed of sprouting ability.Persistently seed bank is to plant
Thing adapts to the important Reproductive Strategy in the particularly severe habitat of environment, for understanding the interaction of plant and environment, plant species
Adapt to significant with evolution.
Plant develops the breeding plan that multiple Reproductive Strategy, lasting seed bank not all plant share during evolution
Slightly.Diffusion and the sprouting of seed can be completed then, it is not necessary to the most there is not lasting seed bank as elm utilizes wind-force to propagate.Cause
This, it is relevant which plant has lasting seed bank and its local environment, its own evolution strategy.The research lasting seed bank of plant is reason
Solve basis and the premise of plant propagation strategy.
The most persistently seed bank research relies primarily on physical partition method and seed germination method.Physical partition method is application water drift
Wash, sieve screens and in anatomical lens or the method such as basis of microscopic observation separation, first seed is picked out from soil, logical
Cross qualification Bamboo resource, determine the method whether species have lasting seed bank.The method must include identifying seed vitality
Step, otherwise cannot judge that seed has sprouting ability, and common method has teterazelium test and Direct Test embryo method.Seed germination
Method is exactly seed sample in soil to be placed under applicable temperature, moisture content and illumination condition sprouts, and is sprouted by qualification
The kind of seedling, determines whether species have lasting seed bank.Visible, the most persistently judgement of seed bank needs to set up in experiment
On Research foundation, there is the shortcomings such as time-consuming, laborious.
Therefore, set up a kind of Forecasting Methodology judged for lasting seed bank, for lasting seed bank and correlational study, reason
Solve the plant response mechanism to environment, promote that revegetation and species diversity conservation are significant.
Summary of the invention
It is an object of the invention to depend on unduly the spy of time-consuming, laborious experimental technique for the research of current persistent seed bank
Point, proposes a kind of lasting seed bank Forecasting Methodology based on random forest, effectively to study the lasting seed bank of plant.
Based on random forest the lasting seed bank Forecasting Methodology that the present invention provides, its method is as follows:
1) collect plant seed character, build seed properties data base；
2) collect plant lasting seed bank attribute, build lasting seed bank data base；
3) using species name as associations, seed properties data base is associated with lasting seed bank data base；
4) from seed properties data base and lasting seed database, randomly select data, delete and only have seed properties or persistently
The data of seed bank information, utilize random forests algorithm to build training set；
5) based on described training set, set up forecast model, it was predicted that model has multiple decision tree, throw according to the classification of decision tree
Ticket result, it is achieved classification；
6) choose from seed properties and lasting seed bank linked database and do not comprise the data of lasting seed bank information, set up pre-
Survey collection；
7) utilize described forecast model, forecast set be predicted, obtain lasting seed bank in the presence/absence of probit；
8) according to probability size, lasting seed bank is classified as the class that probability is bigger, i.e. can determine whether whether corresponding plant exists and hold
Seed bank for a long time.
In described step 3), seed properties data base is species with the Main Basis of lasting seed bank database association
Latin name.
In training set described in step 4), the seed properties that same packet contains is no less than 2.
The invention has the beneficial effects as follows:
(1) utilize random forests algorithm to simplify or save a large amount of manpowers and time needed in the research of lasting seed bank, having
Help improve Efficiency, cost-effective.
(2) cannot confirm, for the time, the species whether lasting seed bank exists, such as the index such as weight, shape can
On the basis of survey, preliminary judgement lasting seed bank situation can be carried out；Extremely rare difficult for newfound plant species or seed
To carry out the plant species of experiment, the method has important value.
Detailed description of the invention
Lasting seed bank Forecasting Methodology based on random forest, step is as follows:
1) collect plant seed character, build seed properties data base；
2) collect plant lasting seed bank attribute, build lasting seed bank data base；
3) using species name as associations, seed properties data base is associated with lasting seed bank data base；
4) from seed properties data base and lasting seed database, randomly select data, delete and only have seed properties or persistently
The data of seed bank information, utilize random forests algorithm to build training set；
5) based on described training set, set up forecast model, it was predicted that model has multiple decision tree, throw according to the classification of decision tree
Ticket result, it is achieved classification；
6) choose from seed properties and lasting seed bank linked database and do not comprise the data of lasting seed bank information, set up pre-
Survey collection；
7) utilize described forecast model, forecast set be predicted, obtain lasting seed bank in the presence/absence of probit；
8) according to probability size, lasting seed bank is classified as a class of probability relatively big (probability is more than 50%), i.e. can determine whether to plant accordingly
Whether thing exists lasting seed bank.
Wherein, in step 3), the Main Basis of seed properties data base and lasting seed bank database association is that species draw
Fourth name.
In training set described in step 4), the seed properties that same packet contains is no less than 2.
The inventive method is particularly applicable in Keerqin sandy land Activities of Some Plants lasting seed bank research process as follows:
The lasting seed bank of application examples Keerqin sandy land Activities of Some Plants is studied
Should mainly use the Keerqin sandy land Activities of Some Plants kind data of collection by use-case, of the present invention application process is expanded on further,
Specifically comprise the following steps that
1. collect seed properties, build seed properties data base
We collect the data such as the seed weight of 143 kinds of plants of Keerqin sandy land, shape of the seed, accessory structure, living form.
2. collect plant lasting seed bank attribute, build lasting seed bank data base
We collect whether 91 kinds of plant seeds of Keerqin sandy land have lasting seed bank information.Wherein there is lasting seed bank
Plant 52 kinds, not there is the plant 39 kinds of lasting seed bank.
3. seed properties data base is associated with lasting seed bank data base
Compareing through species and compare, and deleting partial invalidity information, we only remain has the 91 of lasting seed bank data
Individual species information.Calculating for simplifying, we only retain seed weight and the data of shape of the seed two indices.Seed weight is maximum
Value is 130.8 mg, and minima is that 0.05 mg, median and average are respectively 0.99 mg and 5.62 mg.Shape of the seed is maximum
Value is 0.207, and minima is 0.007, and median and average are respectively 0.088 and 0.093.
4. utilize random forests algorithm to build training set
Owing to random forest method need not cross validation, we select 90 species therein as training set, remain 1 thing
Kind Artemisia halodendron (Artemisia halodendron) for verifying the accuracy of prediction.
5., based on described training set, set up forecast model
According to random forests algorithm, we establish forecast model.This model employs 500 decision trees, and 00B error rate is
32.97%。
6. set up forecast set
In this example, it was predicted that collection only have species, i.e. Artemisia halodendron (Artemisia halodendron), its seed weight is
0.51 mg, shape of the seed is 0.12, and these species have lasting seed bank.
7. utilize described forecast model, it was predicted that collection is predicted
The forecast model that we set up before utilizing is predicted, and obtain these species having the probability of lasting seed bank is 87%, no
The probability with lasting seed bank is 13%.Can be judged as that there is lasting seed bank.And this judgement meets reality
Situation.
8. according to predicting the outcome, it may be judged whether whether corresponding plant exists lasting seed bank.
According to prediction probability, we are judged as having lasting seed bank.And this judgement tallies with the actual situation.
Claims (3)
1. lasting seed bank Forecasting Methodology based on random forest, it is characterised in that step is as follows:
Collect plant seed character, build seed properties data base；
Collect plant lasting seed bank attribute, build lasting seed bank data base；
Using species name as associations, seed properties data base is associated with lasting seed bank data base；
From seed properties data base and lasting seed database, randomly select data, delete and only have seed properties or persistently plant
The data of word bank information, utilize random forests algorithm to build training set；
Based on described training set, set up forecast model, it was predicted that model has multiple decision tree, vote according to the classification of decision tree
Result, it is achieved classification；
From seed properties with lasting seed bank linked database, choose the data not comprising lasting seed bank information, set up prediction
Collection；
Utilize described forecast model, forecast set be predicted, obtain lasting seed bank in the presence/absence of probit；
According to probability size, lasting seed bank is classified as the class that probability is bigger, i.e. can determine whether whether corresponding plant exists persistently
Seed bank.
2. lasting seed bank Forecasting Methodology based on random forest as claimed in claim 1, it is characterised in that described step
3) in, seed properties data base is species latin name with the Main Basis of lasting seed bank database association.
3. lasting seed bank Forecasting Methodology based on random forest as claimed in claim 1, it is characterised in that described in step 4)
Training set in, the seed properties that same packet contains be no less than 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610659724.6A CN106295887A (en) | 2016-08-12 | 2016-08-12 | Lasting seed bank Forecasting Methodology based on random forest |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610659724.6A CN106295887A (en) | 2016-08-12 | 2016-08-12 | Lasting seed bank Forecasting Methodology based on random forest |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106295887A true CN106295887A (en) | 2017-01-04 |
Family
ID=57668758
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610659724.6A Pending CN106295887A (en) | 2016-08-12 | 2016-08-12 | Lasting seed bank Forecasting Methodology based on random forest |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106295887A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109214609A (en) * | 2018-11-15 | 2019-01-15 | 辽宁大学 | A kind of Prediction of annual electricity consumption method based on fractional order discrete grey model |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104346513A (en) * | 2013-08-09 | 2015-02-11 | 苏州润新生物科技有限公司 | Chinese herbal medicinal ingredient and compound hepatotoxin evaluation system based on propelling decision-making tree |
| CN104615910A (en) * | 2014-12-30 | 2015-05-13 | 中国科学院深圳先进技术研究院 | Method for predicating helix interactive relationship of alpha transmembrane protein based on random forest |
| US20150167085A1 (en) * | 2013-09-09 | 2015-06-18 | Daniel R. Salomon | Methods and Systems for Analysis of Organ Transplantation |
| CN105577660A (en) * | 2015-12-22 | 2016-05-11 | 国家电网公司 | DGA domain name detection method based on random forest |
-
2016
- 2016-08-12 CN CN201610659724.6A patent/CN106295887A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104346513A (en) * | 2013-08-09 | 2015-02-11 | 苏州润新生物科技有限公司 | Chinese herbal medicinal ingredient and compound hepatotoxin evaluation system based on propelling decision-making tree |
| US20150167085A1 (en) * | 2013-09-09 | 2015-06-18 | Daniel R. Salomon | Methods and Systems for Analysis of Organ Transplantation |
| CN104615910A (en) * | 2014-12-30 | 2015-05-13 | 中国科学院深圳先进技术研究院 | Method for predicating helix interactive relationship of alpha transmembrane protein based on random forest |
| CN105577660A (en) * | 2015-12-22 | 2016-05-11 | 国家电网公司 | DGA domain name detection method based on random forest |
Non-Patent Citations (1)
| Title |
|---|
| 张雷、王琳琳、张旭东、刘世荣、孙鹏森、王同立: "随机森林算法基本思想及其在生态学中的应用——以云南松分布模拟为例", 《生态学报》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109214609A (en) * | 2018-11-15 | 2019-01-15 | 辽宁大学 | A kind of Prediction of annual electricity consumption method based on fractional order discrete grey model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Farrior et al. | Competition for water and light in closed-canopy forests: a tractable model of carbon allocation with implications for carbon sinks | |
| Phillips et al. | A belowground perspective on the drought sensitivity of forests: Towards improved understanding and simulation | |
| Potvin et al. | Biodiversity enhances individual performance but does not affect survivorship in tropical trees | |
| Poorter et al. | Are functional traits good predictors of demographic rates? Evidence from five neotropical forests | |
| Verkerk et al. | Assessing impacts of intensified biomass removal on deadwood in European forests | |
| Nouvellon et al. | Within-stand and seasonal variations of specific leaf area in a clonal Eucalyptus plantation in the Republic of Congo | |
| Du et al. | Phylogenetic constraints and trait correlates of flowering phenology in the angiosperm flora of C hina | |
| Jamoneau et al. | Fragmentation alters beta‐diversity patterns of habitat specialists within forest metacommunities | |
| McNicol et al. | How resilient are African woodlands to disturbance from shifting cultivation? | |
| Mekuria et al. | Changes in woody species composition following establishing exclosures on grazing lands in the lowlands of Northern Ethiopia | |
| Goodale et al. | Disturbance and tropical pioneer species: Patterns of association across life history stages | |
| Pasztor et al. | Drivers of the bark beetle disturbance regime in Alpine forests in Austria | |
| Reitalu et al. | Small-scale plant species richness and evenness in semi-natural grasslands respond differently to habitat fragmentation | |
| Douglas et al. | Reducing shallow landslide occurrence in pastoral hill country using wide‐spaced trees | |
| Htun et al. | Tree species composition and diversity at different levels of disturbance in Popa Mountain Park, Myanmar | |
| Arellano et al. | Crown damage and the mortality of tropical trees | |
| De Ridder et al. | Tree-ring analysis of an African long-lived pioneer species as a tool for sustainable forest management | |
| Tullus et al. | Early growth of hybrid aspen (Populus× wettsteinii Hämet-Ahti) plantations on former agricultural lands in Estonia | |
| Lohmus et al. | Long‐term colonization ecology of forest‐dwelling species in a fragmented rural landscape–dispersal versus establishment | |
| Sverdrup-Thygeson et al. | Do conservation measures in forest work? A comparison of three area-based conservation tools for wood-living species in boreal forests | |
| Stark et al. | Age and sex‐specific rates of leaf regeneration in the Mojave Desert moss Syntrichia caninervis | |
| Condé et al. | Effects of sustainable forest management on tree diversity, timber volumes, and carbon stocks in an ecotone forest in the northern Brazilian Amazon | |
| Zhang et al. | Factors affecting the distribution and abundance of Asplenium nidus L. in a tropical lowland rain forest in Peninsular Malaysia | |
| Pickles et al. | Host and habitat filtering in seedling root-associated fungal communities: taxonomic and functional diversity are altered in ‘novel’soils | |
| Rutherford et al. | Seedling response to environmental variability: The relationship between phenotypic plasticity and evolutionary history in closely related Eucalyptus species |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170104 |
|
| RJ01 | Rejection of invention patent application after publication |