CN110287423A - A kind of farm Products Show system and method based on collaborative filtering - Google Patents
A kind of farm Products Show system and method based on collaborative filtering Download PDFInfo
- Publication number
- CN110287423A CN110287423A CN201910377393.0A CN201910377393A CN110287423A CN 110287423 A CN110287423 A CN 110287423A CN 201910377393 A CN201910377393 A CN 201910377393A CN 110287423 A CN110287423 A CN 110287423A
- Authority
- CN
- China
- Prior art keywords
- user
- farm
- product
- collaborative filtering
- scoring
- 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
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000001914 filtration Methods 0.000 title claims abstract description 23
- 239000011159 matrix material Substances 0.000 claims abstract description 28
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 238000011156 evaluation Methods 0.000 claims abstract description 5
- 230000003542 behavioural effect Effects 0.000 claims abstract description 4
- 238000011002 quantification Methods 0.000 claims description 5
- 239000002917 insecticide Substances 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 239000007921 spray Substances 0.000 claims description 3
- 238000009333 weeding Methods 0.000 claims description 2
- 210000003813 thumb Anatomy 0.000 claims 1
- 238000009423 ventilation Methods 0.000 claims 1
- 241000196324 Embryophyta Species 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 244000025254 Cannabis sativa Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000011524 similarity measure Methods 0.000 description 1
- 235000013311 vegetables Nutrition 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- 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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Agronomy & Crop Science (AREA)
- Animal Husbandry (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Mining & Mineral Resources (AREA)
- Marine Sciences & Fisheries (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- General Engineering & Computer Science (AREA)
Abstract
The invention discloses a kind of farm Products Show system and method based on collaborative filtering, specific step is as follows: step 1: typing farm user's history behavioral data, step 2: the rating matrix of farm user and product is established, step 3: the similarity of user is calculated, step 4: user is grouped using preset sorting algorithm, step 5: the appraisal result for the product that user does not score in different groups is predicted according to the similarity of different user, step 6: recommendation output is carried out according to appraisal result.Pass through the recommender system, the item that do not score user carries out score in predicting, to efficiently solve the low problem of the imperfect caused recommendation precision of user's evaluation in farm game, the invention calculation method is simple simultaneously, step is few, and complexity is small, and counting accuracy is high, the speed and accuracy for recommending specific cultivation or plantation product to farm user are effectively raised, the experience sense of user is improved.
Description
Technical field
The present invention relates to precision Products Show correlative technology field, specially a kind of farm product based on collaborative filtering
Recommender system and method.
Background technique
Today's society, with the fast development of electric business industry, before on popular line virtual farm it is impossible to meet
Contemporary Man's Demands, contemporary citizen have been not content with to plant vegetables in virtual farm, " dish steathily ", these virtual operations, so
The mode of on-line off-line combination based on virtual farm and real farm, which has begun, to come into vogue.This mode is meeting city
While the entertainment requirements of the people, user is also met for the demand of high quality agricultural product by dispatching under line.
The agricultural product recommended method of the mode combined at present for virtual farm and real farm is seldom, in farm
How a large amount of cultivation or varieties of plant information screen according to different users, filter out its interested information, accomplishing essence
Standardization recommended products enhances the use feeling and experience sense of user so that operating on the line of user more efficiently, is the current field
Recommender system need solve, optimization key problem.
Currently, most commonly used recommended method is collaborative filtering recommending method, its core of project-based collaborative filtering recommending
Thought wants the similar article of article for recommending those to like before with them to user, but project-based collaborative filtering pushes away
The similarity that article is not calculated using the contents attribute of article is recommended, the behavior record that it mainly passes through analysis user calculates article
Similarity.
CN108389113A discloses a kind of recommended method and system, according to the similarity of project scoring and item label
Similarity is merged, and new item similarity measure method is had devised.
CN104751353A discloses the collaborative filtering method based on cluster and slope one prediction, and it is similar to introduce attribute
Property improve slope one prediction model, and with user cluster combine applied to collaborative filtering recommending.
However, further investigations have shown that, above-mentioned collaborative filtering recommending does not distinguish the otherness between user, only
The single feedback information for having considered user's evaluation, it is known that scoring record be it is seldom, be easy to lead because of data sparsity problem
User experience is caused to reduce, recommendation effect is not fine.
Summary of the invention
The purpose of the present invention is to provide a kind of farm Products Show system and method based on collaborative filtering, on solving
State the problem of proposing in background technique.
To achieve the above object, the invention provides the following technical scheme: a kind of farm Products Show based on collaborative filtering
System and method, farm Products Show system and method for this kind based on collaborative filtering includes following basic step:
Step 1, typing farm user data: the user data includes a large amount of farm user's history behavioral data.
Step 2 generates rating matrix: the farm user data of processing institute's typing generates the scoring of user-farm product
Matrix, the rating matrix are expressed as R (m, n), wherein m is farm number of users, and n is farm product number.
Step 3 calculates user's similarity: calculating separately the similarity sim (x, y) between different user.
Step 4, user grouping: being grouped all users using preset sorting algorithm, can be more accurately to user
Classify.
Step 5 predicts scoring of the user to the farm product not scored: being calculated using the following equation user x to not scoring
The scoring of product j
Wherein, sim (x, y) indicates the similitude between farm user x and farm user y,WithRespectively indicate user x
The average value to score with user y product, N (x) indicate the nearest-neighbors set of the user x, yjIndicate user y to product j
Scoring.
Step 6 recommends output: predicting user according to step 5 and carries out to the appraisal result for the farm product not scored
The farm Products Show of precision.
Preferably, it also includes the following specific steps: in step 2, when user x does not have score data to farm product j
When, then the scoring is added in x (j)=0, the scoring x (j) that user x in the step 5 predicts the product j that do not score
Matrix.
Preferably, the calculation formula of the similarity between the different user x and y are as follows:
Wherein xi, yiThe scoring of user x and user y to product i is respectively indicated,WithRespectively indicate user x and user y
To the average value of product scoring.
Preferably, the step 2 generates in rating matrix, by the farm user behavior data of typing in the step 1
Specific score value is converted to by quantification rule, the quantification rule includes by a variety of different user experience knots
Fruit is identified with certain numerical value.For example there is common user behavior in farm, login checks crops, watering, fertilising, removes
Grass sprays insecticide, shares, thumbing up, commenting on, evaluation etc. of making a blueprint, and a weight can be arranged to every kind of behavior here, for example be arranged
Proportion range is 1 to 10, and suitable weight is arranged to above-mentioned user behavior according to the actual situation, and final weighting obtains user-production
Judge sub-matrix.
Preferably, this system further includes real-time update user's rating matrix, due to customer data base can increase newly it is some new
Browsing and consumption data, constantly adjust user's rating matrix according to newly-increased database information, to realize that dynamic adjusts
Process is simultaneously supplied to user-customized recommended service.
Compared with prior art, the beneficial effects of the present invention are:
User is grouped by using preset sorting algorithm, can more accurately be distinguished between high similar users
Otherness improves the accuracy for recommending farm product;Calculation method is simple, and step is few, and complexity is small, and the speed of recommendation can be improved
Degree;Dynamic real-time update user-product rating matrix focuses on the otherness between different time sections same subscriber, more human nature
Change, promotes user experience.
Detailed description of the invention
Fig. 1 is a kind of overall procedure schematic diagram of farm recommender system based on collaborative filtering;
Fig. 2 is farm user-product rating matrix schematic diagram.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Referring to Fig. 1, the present invention provides a kind of technical solution: a kind of farm Products Show system based on collaborative filtering and
Method, comprising the following specific steps
Step 1, typing farm user data: the user data includes a large amount of farm user's history behavioral data.
Step 2 generates rating matrix: the farm user data of processing institute's typing generates the scoring of user-farm product
Matrix, the rating matrix are expressed as R (m, n), and wherein m is farm number of users, and n is farm product number.In this step, by institute
The farm user behavior data for stating typing in step 1 is converted to specific score value by quantification rule, and described quantifies
Changing rule includes being identified a variety of different user experience results with certain numerical value.Such as common user's row in farm
To have, login checks crops, watering, fertilising, weeding, sprays insecticide, shares, thumbing up, commenting on, evaluation etc. of making a blueprint, and here may be used
A weight is arranged to every kind of behavior, for example setting proportion range is 1 to 10, is set according to the actual situation to above-mentioned user behavior
Suitable weight is set, final weighting obtains user-product rating matrix.This rating matrix is relative to traditional rating matrix number
According to more comprehensively.
Step 3 calculates user's similarity: calculating separately the similarity sim (x, y) between different user.
Step 4, user grouping: according to the similitude between the user using preset sorting algorithm to all users
It is grouped.In embodiments herein, all users can be grouped using AP (neighbour's propagation) clustering algorithm,
The detailed process of AP clustering algorithm are as follows:
Similarity matrix is generated according to the similarity between different user calculated in step 3, updates similarity matrix
In each point Attraction Degree information, calculate degree of membership information;
Degree of membership information is updated, Attraction Degree information is calculated;
Attraction Degree information and the summation of degree of membership information to sample point, detect its decision for selecting cluster centre;
If its cluster centre is constant after iteration several times or the number of iterations be more than set number or
The decision about sample point in one sub-regions remains unchanged after iteration for several times, then algorithm terminates.
It is possible thereby to which all users are grouped according to the similitude between user, the user in each grouping is high phase
Like property user.
Step 5 predicts scoring of the user to the farm product not scored: being calculated using the following equation user x to not scoring
The scoring of product j
Wherein, sim (x, y) indicates the similitude between farm user x and farm user y,WithRespectively indicate user x
The average value to score with user y product, N (x) indicate the nearest-neighbors set of the user x, yjIndicate user y to product j
Scoring,.
Step 6 recommends output: predicting user according to step 5 and go forward side by side to the appraisal result for the farm product not scored
The farm Products Show of row precision.
Farm recommender system and method based on collaborative filtering, which is characterized in that also include the following specific steps: in step
In two, when user x does not have score data to farm product j, then x (j)=0, by user x in the step 5 to production of not scoring
The rating matrix is added in the scoring x (j) that product j is predicted.
Farm recommender system and method based on collaborative filtering, which is characterized in that the phase between the different user x and y
Like the calculation formula of degree are as follows:
Wherein xi, yiThe scoring of user x and user y to product i is respectively indicated,WithRespectively indicate user x and user y
To the average value of product scoring.
Fig. 2 is embodiment user-product rating matrix partial content, and NULL indicates that user does not score to the product,
Every a line represents the behavior vector of a user.
It although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
A variety of variations, modification, replacement can be carried out to these embodiments without departing from the principles and spirit of the present invention by understanding
And modification, the scope of the present invention is defined by the appended.
Claims (5)
1. a kind of farm Products Show system and method based on collaborative filtering, it is characterised in that: this kind is based on collaborative filtering
Farm Products Show system and method includes following basic step:
Step 1, typing farm user data: the user data includes a large amount of farm user's history behavioral data.
Step 2 generates rating matrix: the farm user data of processing institute's typing generates user-farm product scoring square
Battle array, the rating matrix are expressed as R (m, n), wherein m is farm number of users, and n is farm product number.
Step 3 calculates user's similarity: calculating separately the similarity sim (x, y) between different user.
Step 4, user grouping: being grouped all users using preset sorting algorithm, can more accurately carry out to user
Classification.
Step 5 predicts scoring of the user to the farm product not scored: being calculated using the following equation user x to the product that do not score
The scoring of j
Wherein, sim (x, y) indicates the similitude between farm user x and farm user y,WithRespectively indicate user x and use
The average value that family y scores to product, N (x) indicate the nearest-neighbors set of the user x, yjIndicate that user y comments product j
Point.
Step 6 recommends output: predicting user according to step 5 and carries out precisely to the appraisal result for the farm product not scored
The farm Products Show of change.
2. a kind of farm Products Show system and method based on collaborative filtering according to claim 1, it is characterised in that:
It also includes the following specific steps: in step 2, when user x does not have score data to farm product j, then x (j)=0, by institute
The scoring x (j) that user x predicts the product j that do not score in step 5 is stated, the rating matrix is added.
3. a kind of farm Products Show system and method based on collaborative filtering according to claim 1, it is characterised in that:
The calculation formula of similarity between the different user x and y are as follows:
Wherein xi, yiThe scoring of user x and user y to product i is respectively indicated,WithUser x and user y are respectively indicated to product
The average value of scoring.
4. a kind of farm Products Show system and method based on collaborative filtering according to claim 1, it is characterised in that:
The step 2 generates in rating matrix, and the farm user behavior data of typing in the step 1 is turned by quantification rule
Be changed to specific score value, the quantification rule include by a variety of different user experience results with certain numerical value into
Line identifier.For example there is common user behavior in farm, login checks crops, watering, fertilising, weeding, sprays insecticide, divides
It enjoys, thumb up, commenting on, evaluation etc. of making a blueprint, a weight every kind of behavior can be arranged here, for example setting proportion range arrives for 1
10, suitable weight is arranged to above-mentioned user behavior according to the actual situation, final weighting obtains user-product rating matrix.
5. a kind of farm Products Show system and method based on collaborative filtering according to claim 1, it is characterised in that:
This system further includes real-time update user's rating matrix, since customer data base can increase some new browsings and consumption data newly,
User's rating matrix is constantly adjusted according to newly-increased database information, to realize dynamic adjustment process and be supplied to user
Personalized ventilation system.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910377393.0A CN110287423A (en) | 2019-05-05 | 2019-05-05 | A kind of farm Products Show system and method based on collaborative filtering |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910377393.0A CN110287423A (en) | 2019-05-05 | 2019-05-05 | A kind of farm Products Show system and method based on collaborative filtering |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110287423A true CN110287423A (en) | 2019-09-27 |
Family
ID=68002629
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910377393.0A Pending CN110287423A (en) | 2019-05-05 | 2019-05-05 | A kind of farm Products Show system and method based on collaborative filtering |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110287423A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110889045A (en) * | 2019-10-12 | 2020-03-17 | 平安科技(深圳)有限公司 | Label analysis method, device and computer readable storage medium |
CN111353864A (en) * | 2020-03-31 | 2020-06-30 | 中国建设银行股份有限公司 | Product recommendation method and device, server and storage medium |
CN113822738A (en) * | 2021-06-22 | 2021-12-21 | 昆明理工大学 | Multi-dimensional agricultural product supply and demand bidirectional personalized recommendation method |
-
2019
- 2019-05-05 CN CN201910377393.0A patent/CN110287423A/en active Pending
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110889045A (en) * | 2019-10-12 | 2020-03-17 | 平安科技(深圳)有限公司 | Label analysis method, device and computer readable storage medium |
CN110889045B (en) * | 2019-10-12 | 2024-04-23 | 平安科技(深圳)有限公司 | Label analysis method, device and computer readable storage medium |
CN111353864A (en) * | 2020-03-31 | 2020-06-30 | 中国建设银行股份有限公司 | Product recommendation method and device, server and storage medium |
CN111353864B (en) * | 2020-03-31 | 2024-04-05 | 建信金融科技有限责任公司 | Product recommendation method and device, server and storage medium |
CN113822738A (en) * | 2021-06-22 | 2021-12-21 | 昆明理工大学 | Multi-dimensional agricultural product supply and demand bidirectional personalized recommendation method |
CN113822738B (en) * | 2021-06-22 | 2024-05-14 | 昆明理工大学 | Multi-dimensional agricultural product supply and demand bidirectional personalized recommendation method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wei et al. | Influence of irrigation during the growth stage on yield and quality in mango (Mangifera indica L) | |
Mukherjee et al. | Productivity growth in Indian agriculture: is there evidence of convergence across states? | |
CN110287423A (en) | A kind of farm Products Show system and method based on collaborative filtering | |
CN113229123A (en) | Intelligent irrigation method, device and equipment for crops and storage medium | |
CN104574093B (en) | A kind of method and device that sales volume is calculated based on electric business sample data information | |
Jakopic et al. | It’s great to be the King: Apple fruit development affected by the position in the cluster | |
CN108446351A (en) | The hotel's screening technique and system based on user preference of OTA platforms | |
CN109508407A (en) | The tv product recommended method of time of fusion and Interest Similarity | |
Anderson et al. | Why are US and EU policies toward GMOs so different? | |
Jabir et al. | A strategic analytics using convolutional neural networks for weed identification in sugar beet fields | |
CN108920709A (en) | Based on improving collaborative filtering recommending method under the modified big data of random forest | |
Kao et al. | A modified Roger’s distance algorithm for mixed quantitative–qualitative phenotypes to establish a core collection for Taiwanese vegetable soybeans | |
Bajan et al. | Similarity and Competition of Polish Agri-food Export with the Largest Agricultural Producers in the EU. Analysis of EU, US and China Market | |
CN109005431B (en) | A kind of video evaluations recommender system | |
CN112506977B (en) | Interval intuitionistic fuzzy multi-attribute group decision provider selection method | |
Baihaqi et al. | Coconut farming development strategy in Bireuen Regency using hierarchy process analysis | |
CN112016000B (en) | Movie group recommendation method and system based on convolution collaborative filtering | |
CN109272213A (en) | One kind being based on the digitized vegetable restoring method of the sense of taste | |
Jansen et al. | Towards smarter harvesting from natural palm populations by sparing the individuals that contribute most to population growth or productivity | |
CN113220827A (en) | Construction method and device of agricultural corpus | |
Ansonfino et al. | Diamond model and competition of rubber export markets: Evidence from Sumatra economic growth center | |
Sundari et al. | Analysis of trade potential and factors influencing chili export in Indonesia | |
Stanley | Factors affecting fruit set and fruit quality along branch units of different apricot cultivars | |
Buccola et al. | Optimal market pools for agricultural cooperatives | |
Kunstelj et al. | Using association rules mining for sweet potato (Ipomoea batatas L.) in Slovenia: A case study |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190927 |
|
WD01 | Invention patent application deemed withdrawn after publication |