CN103870604B - Method and apparatus is recommended in tourism - Google Patents
Method and apparatus is recommended in tourism Download PDFInfo
- Publication number
- CN103870604B CN103870604B CN201410136090.7A CN201410136090A CN103870604B CN 103870604 B CN103870604 B CN 103870604B CN 201410136090 A CN201410136090 A CN 201410136090A CN 103870604 B CN103870604 B CN 103870604B
- Authority
- CN
- China
- Prior art keywords
- interest
- point
- model
- user
- represent
- 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 embodiment of the present invention provides a kind of tourism and recommends method and apparatus, the method to include:According to the point of interest Type model and point of interest cost model of setting, the utility function model of point of interest is set up;Object function is generated according to the scoring of the historical interest point of the user and utility function model;With the user preference parameters in object function as optimization aim, optimum utility function model is determined;The value of utility of point of interest to be selected is calculated according to optimum utility function model;At least one maximum point of interest to be selected of value of utility is recommended into the user.The present invention considers the tourism favor of user individual and tourism expense, the tour interest point that accuracy can be recommended higher for user.
Description
Technical field
The present invention relates to travel application field, more particularly to a kind of tourism recommendation method and apparatus.
Background technology
Tourist industry is developed rapidly so as to become one of industry the biggest in the world.According to world tourism and council of travelling
Prediction, from 2011 year 9.1% will bring up to 9.6% to the contribution rate of global GDP to tourist industry in 2021.Online trip is provided for visitor
Trip service becomes numerous tour sites (such as Expedia, the take journey travelling net) trend that develops.However, burgeoning online trip
Trip information selects the sight spot for meeting its individual demand to bring great difficulty to visitor.On the other hand, it is to obtain more industry
Business and profit, tourist enterprise have to be understood that these individual demands and preference of visitor, and improve preferably attractive
Service.Therefore, no matter to visitor or for tourist enterprise, intelligent tourist service all urgently development and raisings.
Existing tourism commending system just with user essential information and each big website to the scoring at sight spot calculating
Similarity between user, is that user recommends sight spot, this tourism commending system to be difficult for using according to the similarity between user
The sight spot of family recommended user satisfaction.
The content of the invention
The invention provides method and apparatus is recommended in a kind of tourism.Consider the tourism favor and tourism expense of user individual
With the tour interest point that accuracy can be recommended higher for user.
The invention provides a kind of tourism recommendation method, including:
According to the point of interest Type model and point of interest cost model of setting, the utility function model of point of interest is set up;
Object function is generated according to the scoring of the historical interest point of user and the utility function model;
With the user preference parameters in the object function as optimization aim, optimum utility function model is determined;
The value of utility of point of interest to be selected is calculated according to optimum utility function model;
At least one maximum point of interest to be selected of the value of utility is recommended into the user.
Present invention also offers a kind of tourism recommendation apparatus, including:
Module is set up, for point of interest Type model and point of interest cost model according to setting, the effect of point of interest is set up
Use function model;
Generation module, generates object function for the historical interest point scoring according to user and the utility function model;
Determining module, for the user preference parameters in the object function as optimization aim, determining optimum utility letter
Exponential model;
Computing module, for the value of utility of point of interest to be selected is calculated according to optimum utility function model;
Recommending module, at least one maximum point of interest to be selected of the value of utility is recommended the user.
Method and apparatus is recommended in a kind of tourism of the invention, by point of interest Type model and point of interest expense according to setting
Model, sets up the utility function model of point of interest;Generated according to the scoring of the historical interest point of user and above-mentioned utility function model
Object function;With the user preference parameters in above-mentioned object function as optimization aim, optimum utility function model is determined;According to upper
State the value of utility that optimum utility function model calculates point of interest to be selected;At least one value of utility maximum point of interest to be selected is recommended
To the user, when point of interest to be selected is recommended for user, it is contemplated that the tourism favor and tourism expense of user individual, can
For the tour interest point that user recommends accuracy higher.
Description of the drawings
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
Accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are these
Some bright embodiments, for those of ordinary skill in the art, without having to pay creative labor, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is the flow chart of present invention tourism recommendation method one embodiment;
Fig. 2 is present invention tourism recommendation method and other realities of tourism recommendation method in terms of the sight spot accuracy rate recommended
Test result figure;
Fig. 3 is present invention tourism recommendation method and other realities of tourism recommendation method in terms of the sight spot error rate recommended
Test result figure;
Fig. 4 be the present invention tourism recommendation method and other tourism recommendation methods recommend sight spot sequence and sight spot most
Experimental result picture between dominating sequence in terms of dependency;
Fig. 5 is the structural representation of present invention tourism recommendation apparatus one embodiment.
Specific embodiment
To make purpose, technical scheme and the advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
The a part of embodiment of the present invention, rather than the embodiment of whole.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
Fig. 1 is the flow chart of present invention tourism recommendation method one embodiment, as shown in figure 1, the execution master of the present embodiment
Body is tourism recommendation apparatus, specifically can be by software, hardware, or the mode that software and hardware combines is realized, then the party
Method includes:
Step 101, according to the point of interest Type model and point of interest cost model of setting, sets up the utility function of point of interest
Model.
In the present embodiment, point of interest can represent the sight spot resource in tourism, such as Great Wall, pyramid, museum etc..Also may be used
To represent the cuisines resource in tourism, such as local characteristic dining room, chafing dish restaurant, western-style restaurant, Japanese cuisine shop etc., point of interest can also
Other resources in travelling are represented, the present embodiment is not restricted.
Specifically, point of interest resource has different types and different expenses, the sight spot money in point of interest represents tourism
During source, expense represents that the admission fee at sight spot is used, and when point of interest represents the cuisines resource in tourism, expense represents cuisines per capita
Consumption, according to the difference of the resource in the tourism that point of interest is represented, the implication that its expense is represented also can be different.With point of interest as trip
As a example by trip sight spot resource, sight spot resource can be divided into geological and geomorphic landscape resources, water landscape resource, biological tourism according to the difference of type
Resource, humane historic site resource, religion culture resource, historic gardens resource etc., and each sight spot resource has different admission fees
With.
In the present embodiment, the point of interest Type model of setting is one with regard to interest vertex type and the user preference of the user
One function of parameter, the preference parameter of user represent that the user values degree, user interest point to every kind of interest vertex type
Cost model is a function of the expense spent with regard to the point of interest of the user, is setting up the utility function model of point of interest
In, when sight spot resource type is more close with user preference, utility function model value is bigger, when the expense of sight spot resource expenditure
More hour, utility function model value is bigger, if utility function model value is bigger, the sight spot resource is more worthwhile for user to push away
Recommend.
Step 102, generates object function according to the scoring of the historical interest point of the user and utility function model.
In the present embodiment, the point of interest that each user went can be different, for different users, gather what the user went
Historical interest point, the user's history point of interest for being gathered include each historical interest point that the user went sight spot type,
Admission fee is used, scoring of the user to each historical interest point, and is scored according to the historical interest point of the user and utility function
Model generates object function, when the scoring of historical interest point and the utility function model value difference more hour of the user, target letter
Number value is less, shows that the utility function model more meets the preference that the user selects point of interest.
Step 103, with the user preference parameters in object function as optimization aim, determines optimum utility function model.
The minimum point of object function in the present embodiment, is asked for, when object function minimalization, shows the utility function
Model more meets the preference that the user selects point of interest, and user preference parameters now are the optimum preference parameter of the user, will
The optimum preference parameter of the user is brought in utility function model, determines the optimum utility function model of the user.
Step 104, calculates the value of utility of point of interest to be selected according to optimum utility function model.
In the present embodiment, point of interest to be selected is the point of interest that the user did not went, by the sight spot type of point of interest to be selected, flower
Expense expense is brought in optimum utility function model, calculates the value of utility of each point of interest to be selected.
At least one maximum point of interest to be selected of value of utility is recommended the user by step 105
Specifically, a maximum point of interest to be selected of value of utility can be recommended user, also can be to be selected emerging by what is calculated
The value of utility of interest point carries out descending arrangement, selects the front K big corresponding point of interest to be selected of value of utility and recommends user, by user
The front K point of interest to be selected to recommending is selected, and customer satisfaction system point of interest to be selected is obtained in user interaction process.
In the present embodiment, by point of interest Type model and point of interest cost model according to setting, point of interest is set up
Utility function model;Object function is generated according to the scoring of the historical interest point of the user and utility function model;With object function
In user preference parameters be optimization aim, determine optimum utility function model;Calculated according to above-mentioned optimum utility function model
The value of utility of point of interest to be selected;At least one maximum point of interest to be selected of value of utility is recommended into the user, is being pushed away for user
When recommending point of interest to be selected, it is contemplated that the tourism favor and tourism expense of user individual, accuracy can be recommended higher for user
Tour interest point.
Further, the effect of point of interest is set up according to the point of interest Type model of setting and the point of interest cost model of setting
With function model, specially:
The utility function model of point of interest is set up using formula (1),
U (X, C)=α * U1(X)+β*U2(C) (1)
Wherein, alpha+beta=1,0 < α <, 1,0 < β < 1, U1(X) point of interest Type model, U are represented2(C) represent point of interest expense
Model, α, β represent point of interest Type model and weight shared by point of interest cost model respectively, and for different users, which is to emerging
The type and expense of interest point has different degrees of concern, so for different users, the value of α, β can be carried out according to user's request
Change.
In the present embodiment, the utility function model of the point of interest of foundation is with regard to point of interest Type model and point of interest expense
Model linear separability, the different users weight different with the setting of expense degree of concern to the type of point of interest, energy can be directed to
Enough more flexible demands for meeting different user.
In economics, Cobb-Douglas function can weigh favorable rating of the consumer to commodity, and form is as follows:
u(y1,y2)=y1 a*y2 b (2)
Wherein, y1And y2The quantity of two class commodity is represented respectively, and a and b describes preference 0 of the consumer to two class commodity
≤ a, b≤1, and a+b=1.
So in the present embodiment, in order to weigh preference of the user to interest vertex type, the point of interest category model of setting
For:
Wherein,X represents interest vertex type vector, X=(x1,x2... ... xi... ..., xn),
xiFor the i-th dimension component of interest vertex type vector X, represent whether point of interest belongs to the i-th type, when point of interest belongs to i-th kind
Type, xiValue is 1, when point of interest is not belonging to the i-th type, xiValue is 0, xi∈ { 0,1 }, when point of interest belong to one or
During multiple different interest vertex types, in the interest vertex type vector represented by X, there are one or more components for 1, work as point of interest
When being not belonging to any one interest vertex type, in the interest vertex type vector represented by X, component value is all 0.αiRepresent user
Preference parameter to i-th kind of interest vertex type, 0≤αi≤ 1, n represent the number of types of point of interest, with the inclined of n interest vertex type
Good parameter constitutes the vector of the n dimensions of an expression user interest type preference parameter for element, and the vector representation is:A=(α1,
α2... ... αi... ..., αn).For different users, represent that the vectorial value of interest pattern preference parameter can be different.
Illustrate by sight spot resource of point of interest, such as sight spot resource according to the difference of type be divided into geological and geomorphic landscape resources,
Water landscape resource, organism tourism resource, humane historic site resource, religion culture resource, historic gardens resource this six classes resource, this
Six class sight spot resources use vector X=(x successively1,x2... ... xi... ..., xn) in one-component represent, if sight spot resource for length
During city, the sight spot resource type on Great Wall is humane historic site resource, then and X=(0,0,0,1,0,0), if the user is only to humanity
A kind of historic site resource this sight spot resource type has preference, then and A=(0,0,0,1,0,0).
In the present embodiment, U1(X) codomain is [0,1].When a certain point of interest is not belonging to any one interest vertex type,
That is (0,0,0,0,0, when 0), point of interest Type model obtains minima, i.e. U to X=1(X)=0, when the affiliated type of point of interest with should
When user preference distribution is matched completely, point of interest Type model obtains maximum, i.e. U1(X)=1.Such as in the example above, Great Wall
The type is matched completely with the user preference distribution, and now point of interest Type model obtains maximum, U1(X)=1。
Further, point of interest cost model is represented by shown in formula (4)
Wherein, C represents the expense of point of interest, CmaxIn the historical interest point that expression user went, most expensive expense, works as interest
The expense of point is higher, and the value of point of interest cost model is less, shows that user selects the probability of the point of interest can be less, works as expense
During with more than a certain special value, the value of point of interest cost model is zero, U2(C) codomain is [0,1], when the flower of point of interest
When expense expense is zero, point of interest cost model value is maximum, when the expense of user effort is the expense that history spends, illustrates this
Point of interest has reached the receptible maximum expense of user, and now point of interest cost model value is minimum.
In the present embodiment, object function is generated according to the scoring of the historical interest point of the user and the utility function model,
Specially:
Object function is generated according to formula (5);
Wherein,M represents the historical interest point number of the user, rjRepresenting should
Scoring of the user to j-th historical interest point, K represent map const.Due to scoring r of the user to historical interest pointjFor in 0-5
Any one integer, the value of utility function model is [0,1], so map const K values are 5.So in formula (5) institute table
In the object function for showing, object function value is less, shows that the utility function model more meets the need that the user selects point of interest
Ask.
Preferably, it is determined that during the optimal value of user preference parameters in object function, using gradient descent method.
The thought of gradient descent method is exactly determining the new direction of search of each iteration using negative gradient direction so that every
Secondary iteration can be such that the object function of the optimization progressively reduces, when the target function value of the optimization of last iteration acquirement changes with this
For rear acquirement optimization target function value difference less than a certain threshold value when, iteration terminates, now corresponding A in the object function
It is the optimal value of the user preference parameters.
Optimum utility function model is determined using formula (6).
Wherein, alpha+beta=1,0 < α <, 1,0 < β < 1, U1(X) point of interest Type model, U are represented2(C) represent point of interest expense
Model, α, β represent point of interest Type model and weight shared by point of interest cost model respectively, and X represents interest vertex type vector, xi
Represent the i-th dimension component of interest vertex type vector X, αiPreference parameter of the user to i-th kind of interest vertex type is represented, n represents emerging
The number of types of interest point.
The optimum user preference parameters are brought in formula (6) the optimum utility function model that can obtain the user,
The user is treated the relevant parameter of the point of interest of selection is brought in optimum utility function, you can obtain the effectiveness of point of interest to be selected
At least one maximum point of interest to be selected of value of utility is recommended the user by value.
Effective result of the present embodiment can be further illustrated by following emulation:
1 emulation content:The sight spot that certain user's history was gone is divided into into training set and test set, in present invention tourism recommendation side
Method (GD) and using particle cluster algorithm calculate user preference parameters method(PSO)In, training set is used for the optimum for determining the user
Utility function model, traditional collaborative filtering tourism recommendation method(CF)In, training set is used to determine with user's similarity most
High other users, so as to the sight spot recommended for the user and its similarity highest other users history was gone.Test set is used
User preference parameters method is calculated in test present invention tourism recommendation method (GD) and using particle cluster algorithm(PSO), it is traditional
Collaborative filtering tourism recommendation method recommends the performance at sight spot good and bad.The recommendation accuracy rate at the sight spot recommended from each method, each method
Comment in terms of dependency three between the sequence at the sight spot that the recommendation error rate at the sight spot of recommendation, each method are recommended and sight spot optimal sequencing
The respective performance of valency these methods.
2 the simulation experiment results
Experimental result of the sight spot that A each methods are recommended in terms of accuracy rate is recommended
The user preference parameters method calculated with present invention tourism recommendation method (GD) and using particle cluster algorithm
(PSO), traditional collaborative filtering tourism recommendation method (CF) is respectively K most attractive point of interest, P before user recommends
Set of user's scoring not less than 4 sight spot in the sight spot for is recommended in expression, and T represents the collection at sight spot in the test set of the user
Close, the recommendation accuracy rate Precision@K at sight spot are expressed as shown in formula (6):
Fig. 2 is present invention tourism recommendation method and other realities of tourism recommendation method in terms of the sight spot accuracy rate recommended
Result figure is tested, X-axis represents the value of point of interest number K of recommendation in fig. 2, vertical coordinate is represented recommends accuracy rate, from Fig. 2
As can be seen that present invention tourism recommendation method (GD) recommends accuracy rate highest, user preference parameters are calculated using particle cluster algorithm
Method (PSO) recommendation accuracy rate slightly below present invention tourism recommendation method (GD), and traditional collaborative filtering tourism recommendation method
(CF)Recommend accuracy rate minimum, so, present invention tourism recommendation method is better than adopting particle cluster algorithm in terms of accuracy rate is recommended
Calculate user preference parameters method (PSO) and traditional collaborative filtering tourism recommendation method (CF).
Experimental result of the sight spot that B each methods are recommended in terms of error rate is recommended
User preference parameters method (PSO) is calculated with present invention tourism recommendation method (GD) and using particle cluster algorithm,
Traditional collaborative filtering tourism recommendation method (CF) is respectively K most attractive point of interest before user recommends, and works as user
Scoring be less than 4 when, it is believed that the user does not like this sight spot, if these sight spots in recommendation results, illustrates recommendation results
It is inaccurate.So recommending its definition as follows:Q is sight spot set of the user's scoring less than 4 in the sight spot recommended, and T is to survey
Examination concentrates the recommendation error rate ErrorRate@K at sight spot set, sight spot to be expressed as shown in formula (7):
Fig. 3 is present invention tourism recommendation method and other experimental results of tourism recommendation method in terms of error rate is recommended
Figure, in Fig. 3, X-axis represents the value of point of interest number K of recommendation, and vertical coordinate is represented recommends error rate, recommends error rate lower
The recommendation sight spot for representing the method more meets the user's request, from figure 3, it can be seen that present invention tourism recommendation method (GD) is pushed away
Recommend error rate minimum, user preference parameters method (PSO) is calculated using particle cluster algorithm and recommends error rate to be slightly above present invention side
Method (GD), and traditional collaborative filtering tourism recommendation method (CF) recommends error rate highest, so, present invention tourism recommendation method
User preference parameters method (PSO) and traditional collaborative filtering trip are calculated better than using particle cluster algorithm in terms of error rate is recommended
Trip recommendation method (CF).
Experimental result between the sequence and sight spot optimal sequence at the sight spot that C each methods are recommended in terms of dependency
User preference parameters method (PSO) is calculated with present invention tourism recommendation method (GD) and using particle cluster algorithm,
Traditional collaborative filtering tourism recommendation method (CF) is respectively K most attractive point of interest before user recommends, and then will
The optimal sequence at the sight spot that the sight spot sequence that each method is recommended is selected with the user is compared, the sight spot sequence that each method is recommended
It is expressed as shown in formula (8) with the dependency NDCG@K of the optimal sequence at sight spot:
Wherein,
Subscripts of the i for sight spot in the sequence of sight spot, riFor scoring of the user to i-th sight spot, iDCG@K be sight spot most
DCG@K values under dominating sequence.
The standard of NDCG@K is based on following two hypothesis:
(1) the higher sight spot of dependency comes above better.
(2) the high sight spot of the dependency sight spot low compared with dependency can more meet the demand of the user.
Fig. 4 be the present invention tourism recommendation method and other tourism recommendation methods recommend sight spot sequence and sight spot most
Experimental result picture between dominating sequence in terms of dependency, in Fig. 4, the post figure of black represents that the sight spot number that each method is recommended is 5
Dependency between sight spot sequence and optimal sequence that sight spot number is 5, white post figure represent that the sight spot number that each method is recommended is
Dependency between 10 sight spot sequence and optimal sequence that sight spot number is 10, the sight spot sequence of recommendation and the optimal sequence at sight spot
Dependency it is bigger, show that sight spot that the method is recommended more meets the demand of the user, figure 4, it is seen that trip of the present invention
In K=5 and K=10, the value of NDCG@K has respectively reached 0.6502 and 0.6649, is above adopting particle for trip recommendation method (GD)
Group's algorithm calculates user preference parameters method (PSO) and traditional collaborative filtering tourism recommendation method (CF) in K=5 and K=10
NDCG@K value, so, the inventive method (GD) recommend sight spot sequence and sight spot optimal sequence between dependency side
Face calculates user preference parameters method (PSO) and traditional collaborative filtering tourism recommendation method better than using particle cluster algorithm
(CF)。
So, present invention tourism recommendation method (GD) is in the recommendation mistake for recommending the recommendation accuracy rate at sight spot, recommend sight spot
User is calculated using particle cluster algorithm better than in terms of dependency three between rate, the sequence at the sight spot recommended and sight spot optimal sequence
Preference parameter method (PSO) and traditional collaborative filtering tourism recommendation method (CF).
Fig. 5 is the structural representation of present invention tourism recommendation apparatus one embodiment, as shown in figure 5, the device can be wrapped
Include:Set up module 501, generation module 502, determining module 503, computing module 504 and recommending module 505.
Wherein, module 501 is set up, for point of interest Type model and point of interest cost model according to setting, sets up emerging
The utility function model of interest point.
Generation module 502, generates object function for the historical interest point scoring according to the user and utility function model.
Determining module 503, for the user preference parameters in object function as optimization aim, determining optimum utility function
Model.
Computing module 504, for the value of utility of point of interest to be selected is calculated according to optimum utility function model.
Recommending module 505, at least one maximum point of interest to be selected of value of utility is recommended the user.
The device of the present embodiment can perform the technical scheme of embodiment of the method shown in Fig. 1, and which realizes principle and technology effect
Seemingly, here is omitted for fruit.
Further, module 501 is set up, for the point of interest Type model according to setting and the point of interest expense mould of setting
Type sets up the utility function model of point of interest, specially:
Using U (X, C)=α * U1(X)+β*U2(C) the utility function model of the point of interest is set up,
Wherein, alpha+beta=1,0 < α <, 1,0 < β < 1, U1(X) point of interest Type model, U are represented2(C) represent point of interest expense
Model, α, β represent point of interest Type model and weight shared by point of interest cost model respectively.
Specifically, point of interest category model is:
Wherein,X represents interest vertex type vector, xiTable
Show the i-th dimension component of interest vertex type vector X, αiPreference parameter of the user to i-th kind of interest vertex type is represented, n represents interest
The number of types of point.
Point of interest cost model is:
Wherein, C represents the expense of point of interest, CmaxRepresent on the user
Most expensive expense in the historical interest point for going.
Generation module 502, generates object function for the historical interest point scoring according to the user and utility function model,
Specially:
According toGenerate the object function;
Wherein,M represents the historical interest point number of the user, rjRepresent
J-th historical interest point scoring, K represent map const, K=5.
Determining module 503, for the user preference parameters in object function as optimization aim, determining optimum utility function
Model, specifically includes:
The optimal value of user preference parameters in the object function is determined using gradient descent method;
According to the optimal value of the user preference parameters, adoptDetermine optimum utility function model.
Wherein, alpha+beta=1,0 < α <, 1,0 < β < 1, U1(X) point of interest Type model, U are represented2(C) represent point of interest expense
Model, α, β represent point of interest Type model and weight shared by point of interest cost model respectively, and X represents interest vertex type vector, xi
Represent the i-th dimension component of interest vertex type vector X, αiPreference parameter of the user to i-th kind of interest vertex type is represented, n represents emerging
The number of types of interest point.
Finally it should be noted that:Various embodiments above only to illustrate technical scheme, rather than a limitation;To the greatest extent
Pipe has been described in detail to the present invention with reference to foregoing embodiments, it will be understood by those within the art that:Its according to
So the technical scheme described in foregoing embodiments can be modified, or which part or all technical characteristic are entered
Row equivalent;And these modifications or replacement, do not make the essence of appropriate technical solution depart from various embodiments of the present invention technology
The scope of scheme.
Claims (10)
1. a kind of tourism recommendation method, it is characterised in that include:
According to the point of interest Type model and point of interest cost model of setting, the utility function model of point of interest is set up;
Object function is generated according to the scoring of the historical interest point of user and the utility function model;
With the user preference parameters in the object function as optimization aim, optimum utility function model is determined;
The value of utility of point of interest to be selected is calculated according to optimum utility function model;
At least one maximum point of interest to be selected of the value of utility is recommended into the user.
2. method according to claim 1, it is characterised in that the point of interest Type model and point of interest according to setting
Cost model sets up the utility function model of point of interest, specially:
Using U (X, C)=α * U1(X)+β*U2(C) set up the utility function model of the point of interest;
Wherein, alpha+beta=1,0 < α <, 1,0 < β < 1, U1(X) the point of interest Type model, U are represented2(C) represent the interest
Point cost model, α, β represent the point of interest Type model and weight shared by the point of interest cost model respectively.
3. method according to claim 2, it is characterised in that the point of interest Type model is:
Wherein,
X represents interest vertex type vector, xiRepresent the i-th dimension component of the interest vertex type vector X, αiRepresent the user couple
The preference parameter of i-th kind of interest vertex type, n represent the number of types of point of interest;
The point of interest cost model is:
Wherein, C represents the expense of point of interest, CmaxRepresent most expensive expense in the historical interest point that the user went.
4. according to the method in claim 2 or 3, it is characterised in that described to be scored according to the historical interest point of user and institute
State utility function model and generate object function, specially:
According toGenerate the object function;
Wherein,M represents the historical interest point number of the user, rjRepresent j-th
Historical interest point scores, and K represents map const.
5. method according to claim 4, it is characterised in that the user preference parameters with the object function are
Optimization aim, determines optimum utility function model, specifically includes:
The optimal value of user preference parameters in the object function is determined using gradient descent method;
According to the optimal value of the user preference parameters, adoptDetermine optimum utility function model;
Wherein, alpha+beta=1,0 < α <, 1,0 < β < 1, U1(X) the point of interest Type model, U are represented2(C) represent the interest
Point cost model, α, β represent the point of interest Type model and weight shared by the point of interest cost model respectively, and X represents institute
State interest vertex type vector, xiRepresent the i-th dimension component of the interest vertex type vector X, αiRepresent that the user is emerging to i-th kind
The preference parameter of interesting vertex type, n represent the number of types of the point of interest.
6. a kind of tourism recommendation apparatus, it is characterised in that include:
Module is set up, for point of interest Type model and point of interest cost model according to setting, the effectiveness letter of point of interest is set up
Exponential model;
Generation module, generates object function for the historical interest point scoring according to user and the utility function model;
Determining module, for the user preference parameters in the object function as optimization aim, determining optimum utility Function Modules
Type;
Computing module, for the value of utility of point of interest to be selected is calculated according to optimum utility function model;
Recommending module, at least one maximum point of interest to be selected of the value of utility is recommended the user.
7. device according to claim 6, it is characterised in that
It is described to set up module, for the effectiveness of point of interest is set up according to the point of interest Type model and point of interest cost model of setting
Function model, specially:
Using U (X, C)=α * U1(X)+β*U2(C) set up the utility function model of the point of interest;
Wherein, alpha+beta=1,0 < α <, 1,0 < β < 1, U1(X) the point of interest Type model, U are represented2(C) represent the interest
Point cost model, α, β represent the point of interest Type model and weight shared by the point of interest cost model respectively.
8. device according to claim 7, it is characterised in that the point of interest category model is:
Wherein,
X represents interest vertex type vector, xiRepresent the i-th dimension component of the interest vertex type vector X, αiRepresent user to i-th kind
The preference parameter of interest vertex type, n represent the number of types of point of interest;
The point of interest cost model is:
Wherein, C represents the expense of point of interest, CmaxMost expensive expense in the point of interest that expression user's history got on.
9. the device according to claim 7 or 8, it is characterised in that
The generation module, generates object function for the historical interest point scoring according to user and the utility function model,
Specially:
According toGenerate the object function;
Wherein,M represents the historical interest point number of the user, rjRepresent j-th
Historical interest point scores, and K represents map const.
10. device according to claim 9, it is characterised in that
The determining module, for the user preference parameters in the object function as optimization aim, determining optimum utility letter
Exponential model, specifically includes:
The optimal value of user preference parameters in the object function is determined using gradient descent method;
According to the optimal value of the user preference parameters, adoptDetermine optimum utility function model;
Wherein, alpha+beta=1,0 < α <, 1,0 < β < 1, U1(X) the point of interest Type model, U are represented2(C) represent the interest
Point cost model, α, β represent the point of interest Type model and weight shared by the point of interest cost model respectively, and X represents institute
State interest vertex type vector, xiRepresent the i-th dimension component of the interest vertex type vector X, αiRepresent that the user is emerging to i-th kind
The preference parameter of interesting vertex type, n represent the number of types of the point of interest.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410136090.7A CN103870604B (en) | 2014-04-04 | 2014-04-04 | Method and apparatus is recommended in tourism |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410136090.7A CN103870604B (en) | 2014-04-04 | 2014-04-04 | Method and apparatus is recommended in tourism |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103870604A CN103870604A (en) | 2014-06-18 |
CN103870604B true CN103870604B (en) | 2017-03-29 |
Family
ID=50909134
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410136090.7A Active CN103870604B (en) | 2014-04-04 | 2014-04-04 | Method and apparatus is recommended in tourism |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103870604B (en) |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104699812A (en) * | 2015-03-23 | 2015-06-10 | 苏州大学张家港工业技术研究院 | Tourism information pushing method and system |
CN106202252A (en) * | 2016-06-29 | 2016-12-07 | 厦门趣处网络科技有限公司 | Method, system are recommended in a kind of trip analyzed based on user emotion |
CN109409916B (en) * | 2017-08-18 | 2021-10-22 | 重庆赫皇科技咨询有限公司 | Tourism marketing system based on big data platform |
CN107545471B (en) * | 2017-09-19 | 2021-06-11 | 北京工业大学 | Big data intelligent recommendation method based on Gaussian mixture |
CN109376294A (en) * | 2018-08-22 | 2019-02-22 | 中国平安人寿保险股份有限公司 | User interest point Weight Determination and device, storage medium and electronic equipment |
CN112149010B (en) * | 2020-11-01 | 2024-05-24 | 云境商务智能研究院南京有限公司 | Group travel route recommendation method based on attention mechanism |
CN112948686B (en) * | 2021-03-25 | 2023-06-16 | 支付宝(杭州)信息技术有限公司 | Position recommendation processing method and device |
CN112989214A (en) * | 2021-05-20 | 2021-06-18 | 湖北游游智游网络科技有限公司 | Tourism information display method and related equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102483827A (en) * | 2009-04-14 | 2012-05-30 | Jda软件集团有限公司 | Travel price optimization (tpo) |
CN103020308A (en) * | 2013-01-07 | 2013-04-03 | 北京趣拿软件科技有限公司 | Method and device for recommending travel strategy project |
CN103020254A (en) * | 2012-12-20 | 2013-04-03 | 北京趣拿信息技术有限公司 | Information recommending method and device |
CN103678429A (en) * | 2012-09-26 | 2014-03-26 | 阿里巴巴集团控股有限公司 | Recommendation method and device of tour routes |
-
2014
- 2014-04-04 CN CN201410136090.7A patent/CN103870604B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102483827A (en) * | 2009-04-14 | 2012-05-30 | Jda软件集团有限公司 | Travel price optimization (tpo) |
CN103678429A (en) * | 2012-09-26 | 2014-03-26 | 阿里巴巴集团控股有限公司 | Recommendation method and device of tour routes |
CN103020254A (en) * | 2012-12-20 | 2013-04-03 | 北京趣拿信息技术有限公司 | Information recommending method and device |
CN103020308A (en) * | 2013-01-07 | 2013-04-03 | 北京趣拿软件科技有限公司 | Method and device for recommending travel strategy project |
Non-Patent Citations (3)
Title |
---|
A travel sequence recommendation approach based on markov model;dalu guo et al;《IADIS international conference on WWW/Internet 2012》;20121021;全文 * |
inter-temporal variation in the travel time and travel cost parameters of transport models;borjesson.m;《transportation》;20140331;第41卷(第2期);全文 * |
基于用户兴趣建模的推荐方法及应用研究;刘淇;《中国博士学位论文全文数据库》;20131015(第10期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN103870604A (en) | 2014-06-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103870604B (en) | Method and apparatus is recommended in tourism | |
Isard et al. | Methods of interregional and regional analysis | |
CN105045818B (en) | A kind of recommendation methods, devices and systems of picture | |
CN105528955B (en) | Generate the method and its device of road network | |
CN108876032A (en) | A kind of data processing method, device, equipment and the system of object addressing | |
CN106204118A (en) | A kind of method and apparatus found for commercial circle | |
CN106681996A (en) | Method and device for determining interest areas and interest points within geographical scope | |
Qian et al. | A self‐sparse generative adversarial network for autonomous early‐stage design of architectural sketches | |
CN110222959B (en) | Urban employment reachability measuring and calculating method and system based on big data | |
CN109409612A (en) | A kind of paths planning method, server and computer storage medium | |
CN106056606A (en) | Image processing method and device | |
CN103336894B (en) | A kind of city plot ratio automatic Mesh Partition Method | |
Brzeziński | Comparison of fractional order derivatives computational accuracy-right hand vs left hand definition | |
CN106021456A (en) | Point-of-interest recommendation method fusing text and geographic information in local synergistic arrangement | |
CN102136133B (en) | A kind of image processing method and image processing apparatus | |
Liu et al. | Horizontal circulation patterns in a large shallow lake: Taihu Lake, China | |
Zhang et al. | FM-based: algorithm research on rural tourism recommendation combining seasonal and distribution features | |
Leilei et al. | Physical education image analysis based on virtual crowd simulation and FPGA | |
CN105913199A (en) | Energy, economy and environment coordination degree calculation method and device based on ellipsoidal model | |
CN106339578A (en) | Multi-strategy integration hospital patient registration recommendation method | |
CN106155998A (en) | A kind of data processing method and device | |
CN109978211A (en) | Prediction technique and device of the flight into departure from port rate | |
CN105162648B (en) | Corporations' detection method based on backbone network extension | |
CN107092617B (en) | Information processing method and device | |
CN105893480A (en) | Information searching method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |