CN104820723B - A kind of vehicle based on owner of cargo's preference inquiry learning recommends sort method online - Google Patents
A kind of vehicle based on owner of cargo's preference inquiry learning recommends sort method online Download PDFInfo
- Publication number
- CN104820723B CN104820723B CN201510282826.6A CN201510282826A CN104820723B CN 104820723 B CN104820723 B CN 104820723B CN 201510282826 A CN201510282826 A CN 201510282826A CN 104820723 B CN104820723 B CN 104820723B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- owner
- cargo
- mrow
- matching degree
- 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
- G06F16/90—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
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of vehicle based on owner of cargo's preference inquiry learning to recommend sort method online, the described method includes:Full connection figure is established according to the correlation attribute information of all vehicles first;Then vehicle having been browsed according to the owner of cargo and interested calculates probability matrix and the expected matrix by cruising vehicle;So as to calculate the matching degree vector for it is expected vehicle by cruising vehicle and the owner of cargo, the corresponding vehicle of each component characterization of the matching degree vector it is expected the matching degree of vehicle with the owner of cargo, finally the vehicle not browsed is ranked up according to matching degree, the above process is repeated, until the owner of cargo chooses desired vehicle.The method of the present invention is updated sequence to subsequently selected vehicle for user, is improved the matching degree between vehicle ranking results and user's expectation, improve user experience, improve vehicle search efficiency by the preference information of on-line study user.
Description
Technical field
The present invention relates to information technology and automation field, more particularly to a kind of vehicle based on owner of cargo's preference inquiry learning
It is online to recommend sort method.
Background technology
Logistics platform establishes the information bridge between car owner and the owner of cargo, and car owner can carry out issue and the goods of information of vehicles
The lookup of thing, the owner of cargo can also carry out the issue of goods information and vehicle is searched.When the owner of cargo carries out vehicle lookup, the owner of cargo can integrate
Consider that many factors such as safety, driver's credibility, traveling state of vehicle, maintenance situation influence, meet even in same
Under conditions of it is required that, selection of the owner of cargo to vehicle also has obvious Preference.But present logistics platform does not provide mostly to be looked into
Ask option or several simple query options are only provided, the expectation for inquiring about rear vehicle ranking results and the owner of cargo is difficult consistent, sequence
As a result it is unsatisfactory;There is presently no the Preference selected according to the owner of cargo vehicle, carry out in vehicle pond other vehicles from
Research in terms of dynamic sequence.How to improve in vehicle pond vehicle sequence and user it is expected between matching degree, realize user preference
On-line study, improve the owner of cargo to the satisfaction of query result, be a technological difficulties urgently to be resolved hurrily.
The content of the invention
It is an object of the invention to overcome, the owner of cargo is when searching vehicle on existing logistics platform, vehicle in existing vehicle pond
The problem of matching degree is not high between sequence and user it is expected, it is proposed that a kind of vehicle based on owner of cargo's preference inquiry learning pushes away online
Sort method is recommended, Preference of this method based on the owner of cargo, can carry out auto-sequencing to vehicle in vehicle pond, improve the owner of cargo and look into
Ask the speed and satisfaction of vehicle.
To achieve these goals, the present invention provides a kind of vehicle based on owner of cargo's preference inquiry learning to recommend to sort online
Method, the described method includes:Full connection figure is established according to the correlation attribute information of all vehicles first;Then according to the owner of cargo
Vehicle browsing and interested calculates probability matrix and the expected matrix by cruising vehicle;So as to calculate not by cruising vehicle with
The owner of cargo it is expected the matching degree vector of vehicle, and the corresponding vehicle of each component characterization of the matching degree vector it is expected vehicle with the owner of cargo
Matching degree, is finally ranked up the vehicle not browsed according to matching degree, the above process is repeated, until the owner of cargo selects
In desired vehicle.
In above-mentioned technical proposal, the method specifically includes following steps:
Step 1) obtains the correlation attribute information of n trolleys to be checked;And establish the full connection figure of vehicle;
Step 2) has browsed m trolleys as the owner of cargo, and clicks on the k platforms checked in m trolleys vehicle interested
When, calculate the expectation value matrix of probability matrix and the owner of cargo to the m trolleys browsed;
Step 3) calculates the matching degree vector that the n-m trolleys not browsed it is expected vehicle with the owner of cargo;
N-m component in the matching degree vector that step 4) obtains step 3) is ranked up from big to small, and according to right
The order answered is ranked up the n-m trolleys not browsed;
If the step 5) owner of cargo does not complete the selection of vehicle, make m=m+1, be transferred to step 3), until the owner of cargo choose it is desired
Vehicle.
In above-mentioned technical proposal, the correlation attribute information in the step 1) includes:The distance of vehicle and the owner of cargo, vehicle carry
Weight, tonnage, cumulative volume, current space available volume, length, service life and service time.
In above-mentioned technical proposal, the detailed process of the full connection figure for establishing vehicle in the step 1) is:
Calculate the connection weight of vehicle node i to vehicle node j:
wij=exp (- αi||xi-xj| |), i=1 ... n, j=1 ... n
Wherein i, j represent node numbering, and each node is a trolley, xi、xjRepresent the attribute arrow of vehicle node i, j
Amount, xi=(xi1,xi2,…xir), xj=(xj1,xj2,…,xjr), r attribute is shared in vector, is represented respectively:Vehicle and the owner of cargo
Distance, vehicle load, tonnage, cumulative volume, current space available volume, length, service life and service time;αi=
(αi1,αi2,…,αir) it is given parameter vector, represent the weight of each attribute, αi1(i=1 ... r) value range is (0,1);
Connection weight according to vehicle node i to vehicle node j, calculates the probability of spreading of vehicle node i to vehicle node j
For:
In above-mentioned technical proposal, the detailed process of the step 2) is:
Probability matrix is:
Wherein, Pm,m=[pij]m×mIt is the probability of spreading p of the m trolleys by being browsedijThe matrix of composition, i=1,2 ...,
M, j=1,2 ..., m;
Pm,n-m=[pij]m×(n-m)It is by the probability of spreading between the m trolleys browsed and the n-m trolleys not browsed
pijForm, i=1,2 ..., m, j=m+1, m+2 ..., n;
Pn-m,mIt is matrix Pm,n-mTransposed matrix;
Pn-m,n-m=[pij](n-m)×(n-m)It is the probability of spreading p of the n-m trolleys by not browsedijThe matrix of composition, i=m+
1, m+2 ..., n, j=m+1, m+2 ..., n;
The expectation value matrix of m trolley of the owner of cargo to browsing is:
Ym=[yi]m×1
Wherein, when i is to click on the sequence number of the k platforms checked vehicle interested, yi=1, otherwise, yi=0.
In above-mentioned technical proposal, the calculating process of the matching degree vector of the step 3) is:
Fn-m=(In-m,n-m-Pn-m,n-m)-1Pn-m,mYm
Wherein, In-m,n-mFor unit matrix;Fn-mFor the matching degree vector of n-m dimensions, each component of the matching degree vector
The corresponding vehicle of characterization it is expected the matching degree of vehicle with the owner of cargo.
The advantage of the invention is that:By the preference information of on-line study user, be user to subsequently selected vehicle into
Row more new sort, improves the matching degree between vehicle ranking results and user's expectation, improves user experience, improve car
Search efficiency.
Brief description of the drawings
Fig. 1 is that the vehicle based on owner of cargo's preference inquiry learning of the present invention recommends the flow chart of sort method online.
Embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.
Step 1) obtains all vehicle correlation attribute informations to be checked;And establish the full connection figure of vehicle;
The sum of all vehicles to be checked is n, and the correlation attribute information of each vehicle includes:Vehicle and the owner of cargo away from
From, vehicle load, tonnage, cumulative volume, current space available volume, length, service life and service time;
Calculate the connection weight of vehicle node i to vehicle node j:
wij=exp (- αi||xi-xj| |), i=1 ... n, j=1 ... n
Wherein i, j represent node numbering, and each node is a trolley, xi、xjRepresent the attribute arrow of vehicle node i, j
Amount, xi=(xi1,xi2,…xir), xj=(xj1,xj2,…,xjr), r attribute is shared in vector, is represented respectively:Vehicle and the owner of cargo
Distance, vehicle load, tonnage, cumulative volume, current space available volume, length, service life and service time;αi=
(αi1,αi2,…,αir) it is given parameter vector, represent the weight of each attribute, αi1(i=1 ... r) value range is (0,1).
Connection weight according to vehicle node i to vehicle node j, calculates the probability of spreading of vehicle node i to vehicle node j
For:
Step 2) has browsed m trolleys as the owner of cargo, and clicks on the k platforms checked in m trolleys vehicle interested
When, calculate the expectation value matrix of probability matrix and the owner of cargo to the m trolleys browsed;
Probability matrix is:
Wherein, Pm,m=[pij]m×mIt is the probability of spreading p of the m trolleys by being browsedijThe matrix of composition, i=1,2 ...,
M, j=1,2 ..., m;
Pm,n-m=[pij]m×(n-m)It is by the probability of spreading between the m trolleys browsed and the n-m trolleys not browsed
pijForm, i=1,2 ..., m, j=m+1, m+2 ..., n;
Pn-m,mIt is matrix Pm,n-mTransposed matrix;
Pn-m,n-m=[pij](n-m)×(n-m)It is the probability of spreading p of the n-m trolleys by not browsedijThe matrix of composition, i=m+
1, m+2 ..., n, j=m+1, m+2 ..., n;
The expectation value matrix of m trolley of the owner of cargo to browsing is:
Ym=[yi]m×1
Wherein, when i is to click on the sequence number of the k platforms checked vehicle interested, yi=1, otherwise, yi=0.
Step 3) calculates the matching degree vector that the vehicle not browsed it is expected vehicle with the owner of cargo:
Fn-m=(In-m,n-m-Pn-m,n-m)-1Pn-m,mYm
Wherein, In-m,n-mFor unit matrix;Fn-mFor the matching degree vector of n-m dimensions, each component of the matching degree vector
The corresponding vehicle of characterization it is expected the matching degree of vehicle with the owner of cargo;
Step 4) is to matching degree vector Fn-mMiddle n-m component is ranked up from big to small, and according to corresponding order to not
The n-m trolleys browsed are ranked up;
If the step 5) owner of cargo does not complete the selection of vehicle, make m=m+1, be transferred to step 3), until the owner of cargo choose it is desired
Vehicle.
Claims (5)
1. a kind of vehicle based on owner of cargo's preference inquiry learning recommends sort method online, the described method includes:
Step S1) full connection figure is established according to the correlation attribute information of all vehicles first;Then browsed according to the owner of cargo
Probability matrix and the expected matrix by cruising vehicle are calculated with vehicle interested;
Step S2) calculate and it is expected that the matching degree of vehicle is vectorial by cruising vehicle and the owner of cargo, each point of the matching degree vector
The corresponding vehicle of scale sign it is expected the matching degree of vehicle with the owner of cargo;Finally the vehicle not browsed is arranged according to matching degree
Sequence,
Step S2 repeatedly), until the owner of cargo chooses desired vehicle;
The method specifically includes:
Step 1) obtains the correlation attribute information of n trolleys to be checked;And establish the full connection figure of vehicle;
Step 2) has browsed m trolleys as the owner of cargo, and when clicking on the k platforms checked in m trolleys vehicle interested, meter
Calculate the expectation value matrix of probability matrix and the owner of cargo to the m trolleys browsed;
Step 3) calculates the matching degree vector that the n-m trolleys not browsed it is expected vehicle with the owner of cargo;
N-m component in the matching degree vector that step 4) obtains step 3) is ranked up from big to small, and according to corresponding
Order is ranked up the n-m trolleys not browsed;
If the step 5) owner of cargo does not complete the selection of vehicle, m=m+1 is made, is transferred to step 3), until the owner of cargo chooses desired car
.
2. the vehicle according to claim 1 based on owner of cargo's preference inquiry learning recommends sort method online, it is characterised in that
Correlation attribute information in the step 1) includes:It is the distance of vehicle and the owner of cargo, vehicle load, tonnage, cumulative volume, currently available
Spatial volume, length, service life and service time.
3. the vehicle according to claim 2 based on owner of cargo's preference inquiry learning recommends sort method online, it is characterised in that
The detailed process of the full connection figure for establishing vehicle in the step 1) is:
Calculate the connection weight of vehicle node i to vehicle node j:
wij=exp (- αi||xi-xj| |), i=1 ... n, j=1 ... n
Wherein i, j represent node numbering, and each node is a trolley, xi、xjRepresent the property vector of vehicle node i, j, xi
=(xi1,xi2,…xir), xj=(xj1,xj2,…,xjr), r attribute is shared in vector, is represented respectively:Vehicle and the owner of cargo away from
From, vehicle load, tonnage, cumulative volume, current space available volume, length, service life and service time;αi=(αi1,
αi2,…,αir) it is given parameter vector, represent the weight of each attribute, αi1(i=1 ... r) value range is (0,1);
Connection weight according to vehicle node i to vehicle node j, the probability of spreading for calculating vehicle node i to vehicle node j are:
<mrow>
<msub>
<mi>p</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<msub>
<mi>w</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>l</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>w</mi>
<mrow>
<mi>i</mi>
<mi>l</mi>
</mrow>
</msub>
</mrow>
</mfrac>
<mo>.</mo>
</mrow>
4. the vehicle according to claim 3 based on owner of cargo's preference inquiry learning recommends sort method online, it is characterised in that
The detailed process of the step 2) is:
Probability matrix is:
<mrow>
<mi>P</mi>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<msub>
<mi>P</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>m</mi>
</mrow>
</msub>
</mtd>
<mtd>
<msub>
<mi>P</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>n</mi>
<mo>-</mo>
<mi>m</mi>
</mrow>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>P</mi>
<mrow>
<mi>n</mi>
<mo>-</mo>
<mi>m</mi>
<mo>,</mo>
<mi>m</mi>
</mrow>
</msub>
</mtd>
<mtd>
<msub>
<mi>p</mi>
<mrow>
<mi>n</mi>
<mo>-</mo>
<mi>m</mi>
<mo>,</mo>
<mi>n</mi>
<mo>-</mo>
<mi>m</mi>
</mrow>
</msub>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
Wherein, Pm,m=[pij]m×mIt is the probability of spreading p of the m trolleys by being browsedijThe matrix of composition, i=1,2 ..., m, j
=1,2 ..., m;
Pm,n-m=[pij]m×(n-m)It is by the probability of spreading p between the m trolleys browsed and the n-m trolleys not browsedijStructure
Into, i=1,2 ..., m, j=m+1, m+2 ..., n;
Pn-m,mIt is matrix Pm,n-mTransposed matrix;
Pn-m,n-m=[pij](n-m)×(n-m)It is the probability of spreading p of the n-m trolleys by not browsedijThe matrix of composition, i=m+1, m+
2 ..., n, j=m+1, m+2 ..., n;
The expectation value matrix of m trolley of the owner of cargo to browsing is:
Ym=[yi]m×1
Wherein, when i is to click on the sequence number of the k platforms checked vehicle interested, yi=1, otherwise, yi=0.
5. the vehicle according to claim 4 based on owner of cargo's preference inquiry learning recommends sort method online, it is characterised in that
The calculating process of the matching degree vector of the step 3) is:
Fn-m=(In-m,n-m-Pn-m,n-m)-1Pn-m,mYm
Wherein, In-m,n-mFor unit matrix;Fn-mFor the matching degree vector of n-m dimensions, each component characterization of the matching degree vector
Corresponding vehicle it is expected the matching degree of vehicle with the owner of cargo.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510282826.6A CN104820723B (en) | 2015-05-28 | 2015-05-28 | A kind of vehicle based on owner of cargo's preference inquiry learning recommends sort method online |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510282826.6A CN104820723B (en) | 2015-05-28 | 2015-05-28 | A kind of vehicle based on owner of cargo's preference inquiry learning recommends sort method online |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104820723A CN104820723A (en) | 2015-08-05 |
CN104820723B true CN104820723B (en) | 2018-05-01 |
Family
ID=53731018
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510282826.6A Active CN104820723B (en) | 2015-05-28 | 2015-05-28 | A kind of vehicle based on owner of cargo's preference inquiry learning recommends sort method online |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104820723B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103336831A (en) * | 2013-07-09 | 2013-10-02 | 清华大学 | Recommendation method and device based on block diagonal matrix |
CN103824192A (en) * | 2012-08-31 | 2014-05-28 | 埃森哲环球服务有限公司 | Hybrid recommendation system |
CN103870581A (en) * | 2014-03-25 | 2014-06-18 | 长沙地大物泊网络科技有限公司 | Car sharing applying and response pairing method and system on basis of centroid dynamic calculation |
CN104331411A (en) * | 2014-09-19 | 2015-02-04 | 华为技术有限公司 | Item recommendation method and item recommendation device |
CN104463548A (en) * | 2014-12-25 | 2015-03-25 | 南京大学 | Carriage quantitative selection method influenced by multiple factors |
CN104657487A (en) * | 2015-03-05 | 2015-05-27 | 东方网力科技股份有限公司 | Licence plate recommendation method and device based on user licence plate querying behavior |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8880439B2 (en) * | 2012-02-27 | 2014-11-04 | Xerox Corporation | Robust Bayesian matrix factorization and recommender systems using same |
US9501762B2 (en) * | 2013-04-23 | 2016-11-22 | Dropbox, Inc. | Application recommendation using automatically synchronized shared folders |
-
2015
- 2015-05-28 CN CN201510282826.6A patent/CN104820723B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103824192A (en) * | 2012-08-31 | 2014-05-28 | 埃森哲环球服务有限公司 | Hybrid recommendation system |
CN103336831A (en) * | 2013-07-09 | 2013-10-02 | 清华大学 | Recommendation method and device based on block diagonal matrix |
CN103870581A (en) * | 2014-03-25 | 2014-06-18 | 长沙地大物泊网络科技有限公司 | Car sharing applying and response pairing method and system on basis of centroid dynamic calculation |
CN104331411A (en) * | 2014-09-19 | 2015-02-04 | 华为技术有限公司 | Item recommendation method and item recommendation device |
CN104463548A (en) * | 2014-12-25 | 2015-03-25 | 南京大学 | Carriage quantitative selection method influenced by multiple factors |
CN104657487A (en) * | 2015-03-05 | 2015-05-27 | 东方网力科技股份有限公司 | Licence plate recommendation method and device based on user licence plate querying behavior |
Also Published As
Publication number | Publication date |
---|---|
CN104820723A (en) | 2015-08-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105829831A (en) | Method for predicting destinations during travel | |
CN103023977B (en) | Recommendation system and method based on positional information | |
Rios-Torres et al. | Fuel consumption for various driving styles in conventional and hybrid electric vehicles: Integrating driving cycle predictions with fuel consumption optimization | |
CN107516235B (en) | Commodity preference estimation method and device | |
US20220101409A1 (en) | Item recommendations weighted by user-valued features | |
CN109409982B (en) | Shared vehicle recommendation method and device | |
Biswas et al. | Selection of commercially available scooters by new MCDM method | |
CN108038629A (en) | A kind of optimization method based on collaborative filtering | |
US20150073933A1 (en) | Vehicle powertrain selector | |
Woody et al. | The role of pickup truck electrification in the decarbonization of light-duty vehicles | |
CN103870507A (en) | Method and device of searching based on category | |
CN109923373A (en) | System and method for determining vehicle reference direction | |
CN111159533A (en) | Intelligent charging service recommendation method and system based on user image | |
CN106156333A (en) | A kind of improvement list class collaborative filtering method of mosaic society information | |
CN111899061B (en) | Order recommendation method, device, equipment and storage medium | |
Limon et al. | Exploration of the drivers influencing the growth of hybrid electric vehicle adoption in the emerging economies: Implications towards sustainability and low-carbon economy | |
CN106021456A (en) | Point-of-interest recommendation method fusing text and geographic information in local synergistic arrangement | |
CN105740327A (en) | Self-adaptive sampling method based on user preferences | |
CN105678335A (en) | Click rate pre-estimation method, device and calculating equipment | |
CN107577786A (en) | A kind of matrix decomposition recommendation method based on joint cluster | |
Liu et al. | Recommending a personalized sequence of pick-up points | |
CN104408189B (en) | The methods of exhibiting and device of keyword ranking | |
CN106126591A (en) | Music data recommends method and system | |
CN104820723B (en) | A kind of vehicle based on owner of cargo's preference inquiry learning recommends sort method online | |
CN106777069A (en) | Personalized recommendation method and commending system based on matrix factorisation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
EXSB | Decision made by sipo to initiate substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |