CN109977131A - A kind of house type matching system - Google Patents
A kind of house type matching system Download PDFInfo
- Publication number
- CN109977131A CN109977131A CN201910266627.4A CN201910266627A CN109977131A CN 109977131 A CN109977131 A CN 109977131A CN 201910266627 A CN201910266627 A CN 201910266627A CN 109977131 A CN109977131 A CN 109977131A
- Authority
- CN
- China
- Prior art keywords
- house type
- matching system
- house
- matched
- algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/12—Hotels or restaurants
Abstract
The invention discloses a kind of house type matching systems, the database including obtaining the house type object composition under system standard hotel；Then house type object and standard flat type object database to be matched, and the integrality of check information are obtained, calculates the similarity of complete house type group, and obtain maximum similarity MX；When MX >=threshold values 0.9, house type group information and MX are saved；Then it returns in matching object database；When MX < threshold values 0.9, mismatch.The present invention can polymerize a large amount of house type in a short time, and full validation has matched house type to reduce erroneous matching, provide the expected polymerization house type of more multiple coincidence user.
Description
Technical field
The present invention relates to hotel service method and technology field, in particular to a kind of house type matching system.
Background technique
With the fast development of the industries such as internet and tourism, hotel industry also flourishes like the mushrooms after rain,
It is both that social development makes so, even more extremely, thus the business of hotel also will be more and more busier for the long feelings institute of people, meanwhile, people
Requirement to hotel service, hotel's house type and scientific management is also higher and higher, and client of today is by pre- on line mostly
Determine hotel, select the house type for oneself being suitble to like, and instantly hotel's house type be it is in different poses and with different expressions, be difficult to by manpower it is exhaustive,
It is many fail to polymerize different types of house type in time be supplied to user and more select, influence the whole benefit of hotel business indirectly
Rate.With the maturation of cloud computing and machine learning, it should be polymerize in conjunction with the technology to polymorphic house type, be verified, user is allowed to have
More more accurately selections.
Summary of the invention
In order to overcome the drawbacks described above of the prior art, the present invention provides a kind of house type matching system, can provide more symbols
Share the expected polymerization house type in family.
The technical scheme adopted by the invention is as follows: a kind of house type matching system, including obtaining the room under system standard hotel
The database of type object composition；Then house type object and standard flat type object database to be matched, and check information are obtained
Integrality, calculate the similarity of complete house type group, and obtain maximum similarity MX；When MX >=threshold values 0.9, house type is saved
Group information and MX；Then it returns in matching object database；When MX < threshold values 0.9, mismatch.
As a preferred solution of the present invention, the method for calculating complete house type group is: including input layer, pretreatment, model
Training and assessment optimization.
As a preferred solution of the present invention, the input layer includes text and data set, and the text is room to be matched
Type object and standard flat type object database；The data include but is not limited to basic training set, verifying collection, test set.
As a preferred solution of the present invention, the pretreatment includes text filtering, former word replacement, enhancing filtering and feature
The building of amount；The text filtering, former word replaces with basic function can be combined by actual conditions in each calculate node；Institute
It states enhancing and is filtered into the power function for carrying out Function Extension by actual scene；The building of the characteristic tensor by feature set house type or
Bed-type, convolution kernel generate characteristic tensor by building method.
As a preferred solution of the present invention, the model training includes classification, cluster and rule；The classification includes linear
With it is non-linear；Described is linearly the solution of spaced planes；It is described it is non-linear generate feature space to promote dimension, then it is flat to interval
The solution in face；The cluster includes division, density and level；It is described to be divided into K-means algorithm；The density is DBSCAN calculation
Method and OPTICS algorithm, and by preferentially being scanned for high density；The level is AgglomerativeClustering calculation
Method；The rule is made of weight cluster and supervision collection, and each level is the supplement or limitation to upper layer.
Compared with prior art, the present invention, which has following technical effect that, utilizes advanced cloud computing ability and suitable business
Neural network, realize house type matching automation and house type relational checking etc..Pass through cheap physical hardware and program software
To simulate artificial and have faster speed progress intensive tasks.It polymerize a large amount of house type in a short time, full validation is
House type is matched to reduce erroneous matching, the expected polymerization house type of more multiple coincidence user is provided.
Detailed description of the invention
Fig. 1 is a kind of flow diagram of house type matching system of the present invention；
Fig. 2 is the flow diagram that the house type matching similarity in a kind of house type matching system of the present invention calculates.
Specific embodiment
Specific embodiments of the present invention will be further explained with reference to the accompanying drawing.It should be noted that for
The explanation of these embodiments is used to help understand the present invention, but and does not constitute a limitation of the invention.In addition, disclosed below
Embodiment of the present invention involved in technical characteristic can be combined with each other as long as they do not conflict with each other.
A kind of house type matching system, the database including obtaining the house type object composition under system standard hotel；Then it obtains
House type object and standard flat type object database to be matched, and the integrality of check information are taken, complete house type group is calculated
Similarity, and obtain maximum similarity MX；When MX >=threshold values 0.9, house type group information and MX are saved；Then matching is returned
In object database；When MX < threshold values 0.9, mismatch.
As a preferred solution of the present invention, the method for calculating complete house type group is: including input layer, pretreatment, model
Training and assessment optimization.
As a preferred solution of the present invention, the input layer includes text and data set, and the text is room to be matched
Type object and standard flat type object database；The data include but is not limited to basic training set, verifying collection, test set.
As a preferred solution of the present invention, the pretreatment includes text filtering, former word replacement, enhancing filtering and feature
The building of amount；The text filtering, former word replaces with basic function can be combined by actual conditions in each calculate node；Institute
It states enhancing and is filtered into the power function for carrying out Function Extension by actual scene；The building of the characteristic tensor by feature set house type or
Bed-type, convolution kernel generate characteristic tensor by building method.
As a preferred solution of the present invention, the model training includes classification, cluster and rule；The classification includes linear
With it is non-linear；Described is linearly the solution of spaced planes；It is described it is non-linear generate feature space to promote dimension, then it is flat to interval
The solution in face；The cluster includes division, density and level；It is described to be divided into K-means algorithm；The density is DBSCAN calculation
Method and OPTICS algorithm, and by preferentially being scanned for high density；The level is AgglomerativeClustering calculation
Method；The rule is made of weight cluster and supervision collection, and each level is the supplement or limitation to upper layer.
Assessment optimization: assessment (objective function [learning objective] reduces the error between calculated result and ideal value)；Optimization
(K-means uses small lot stochastic gradient descent, and SVM uses Lagrangian multiplication and division)；Convolution kernel generates use by largely practicing
Yu Chihua.
Pretreatment:
Text filtering
Some unrelated characters are removed such as to pause character, character of changing voice, western character
Former word replacement
The equivalent word unified standard of word meaning is such as under certain scene:
de luxe->deluxe
Luxurious -> deluxe
economical->economy
economic->economy
Feature Engineering and characteristic tensor construct:
Characteristics dictionary, by grade, type, bed-type, landscape, facility, other etc. divide
The generation of characteristic tensor
Convolution kernel is set
Algorithm design and model training:
Rule-based Nonlinear Classification: consistent from a series of such as strict demand grades, type, landscape, facility need to meet
The rule such as inclusion relation calculates the similarity between house type to be matched and standard flat type.
Matched using neighbor relationships: multiple points are in feature space present position difference, but there are accurate incidence relation,
So one kind can be considered as in [error, distance] to a certain degree.
Detect from cluster: the standard flat type under same standard hotel should have uniqueness, with house type, bed-type two dimensions
There are spacing between the determining each point of degree.
House type matching based on complex network: by similarity combination business that multiple sub-networks are generated by centainly compare into
Row multiple layer combination generates final similarity.
Model evaluation with optimize and revise:
Target:
1 house type to be matched match with most like standard flat type and higher than threshold value 0.9.
2 promote matching rate in the case where guaranteeing accuracy rate.
Continuous optimization:
Characteristics dictionary, the filter capacity of data preprocessing module, the optimization of convolution kernel, the increase of algorithm or complex web network layers
Increase need to keep the robustness of system.
Finally, it should be noted that the foregoing is only a preferred embodiment of the present invention, it is not intended to restrict the invention,
Although the present invention is described in detail referring to the foregoing embodiments, for those skilled in the art, still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features.
All within the spirits and principles of the present invention, any modification, equivalent replacement, improvement and so on should be included in of the invention
Within protection scope.
Claims (5)
1. a kind of house type matching system, it is characterised in that: the data including obtaining the house type object composition under system standard hotel
Library；Then house type object and standard flat type object database to be matched, and the integrality of check information are obtained, is calculated complete
House type group similarity, and obtain maximum similarity MX；When MX >=threshold values 0.9, house type group information and MX are saved；Then
It returns in matching object database；When MX < threshold values 0.9, mismatch.
2. a kind of house type matching system according to claim 1, it is characterised in that: the method for calculating complete house type group
It is: including input layer, pretreatment, model training and assessment optimization.
3. a kind of house type matching system according to claim 2, it is characterised in that: the input layer includes text and data
Collection, the text are house type object and standard flat type object database to be matched；The data are including but not limited to basic
Training set, verifying collection, test set.
4. a kind of house type matching system according to claim 2, it is characterised in that: it is described pretreatment include text filtering,
Former word replacement, enhancing is filtered and the building of characteristic tensor；The text filtering, former word replace with basic function in each calculate node
It can be combined by actual conditions；The enhancing is filtered into the power function that Function Extension is carried out by actual scene；The spy
The building of sign tensor generates characteristic tensor by building method by feature set house type or bed-type, convolution kernel.
5. a kind of house type matching system according to claim 2, it is characterised in that: the model training includes classification, gathers
Class and rule；The classification includes linear and nonlinear；Described is linearly the solution of spaced planes；It is described non-linear for promotion dimension
Degree generates feature space, then the solution to spaced planes；The cluster includes division, density and level；It is described to be divided into K-
Means algorithm；The density is DBSCAN algorithm and OPTICS algorithm, and by preferentially scanning for high density；The layer
Secondary is AgglomerativeClustering algorithm；The rule is made of weight cluster and supervision collection, and each level is to upper layer
Supplement or limitation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910266627.4A CN109977131A (en) | 2019-04-03 | 2019-04-03 | A kind of house type matching system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910266627.4A CN109977131A (en) | 2019-04-03 | 2019-04-03 | A kind of house type matching system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109977131A true CN109977131A (en) | 2019-07-05 |
Family
ID=67082749
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910266627.4A Pending CN109977131A (en) | 2019-04-03 | 2019-04-03 | A kind of house type matching system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109977131A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111340580A (en) * | 2020-02-05 | 2020-06-26 | 深圳市道旅旅游科技股份有限公司 | Method and device for determining house type, computer equipment and storage medium |
| CN113139746A (en) * | 2021-05-13 | 2021-07-20 | 深圳他米科技有限公司 | Hotel stay-continuing method, device, equipment and storage medium based on artificial intelligence |
| CN113628003A (en) * | 2021-07-22 | 2021-11-09 | 上海泛宥信息科技有限公司 | Hotel matching method, system, terminal and storage medium |
-
2019
- 2019-04-03 CN CN201910266627.4A patent/CN109977131A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111340580A (en) * | 2020-02-05 | 2020-06-26 | 深圳市道旅旅游科技股份有限公司 | Method and device for determining house type, computer equipment and storage medium |
| CN113139746A (en) * | 2021-05-13 | 2021-07-20 | 深圳他米科技有限公司 | Hotel stay-continuing method, device, equipment and storage medium based on artificial intelligence |
| CN113139746B (en) * | 2021-05-13 | 2023-11-14 | 深圳他米科技有限公司 | Hotel check-in method, device, equipment and storage medium based on artificial intelligence |
| CN113628003A (en) * | 2021-07-22 | 2021-11-09 | 上海泛宥信息科技有限公司 | Hotel matching method, system, terminal and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113822494B (en) | Risk prediction method, device, equipment and storage medium | |
| WO2021017679A1 (en) | Address information parsing method and apparatus, system and data acquisition method | |
| US20190340533A1 (en) | Systems and methods for preparing data for use by machine learning algorithms | |
| CN108470022B (en) | Intelligent work order quality inspection method based on operation and maintenance management | |
| Nandurge et al. | Analyzing road accident data using machine learning paradigms | |
| CN109635010B (en) | User characteristic and characteristic factor extraction and query method and system | |
| CN104573130A (en) | Entity resolution method based on group calculation and entity resolution device based on group calculation | |
| CN109977131A (en) | A kind of house type matching system | |
| KR102144126B1 (en) | Apparatus and method for providing information for enterprise | |
| CN110310114A (en) | Object classification method, device, server and storage medium | |
| CN110852881A (en) | Risk account identification method and device, electronic equipment and medium | |
| CN113706291A (en) | Fraud risk prediction method, device, equipment and storage medium | |
| CN110825817B (en) | Enterprise suspected association judgment method and system | |
| CN106203165A (en) | The big data analysis method for supporting of information based on credible cloud computing | |
| CN117155771B (en) | Equipment cluster fault tracing method and device based on industrial Internet of things | |
| CN106651461A (en) | Film personalized recommendation method based on gray theory | |
| CN113628043A (en) | Complaint validity judgment method, device, equipment and medium based on data classification | |
| CN111325255B (en) | Specific crowd delineating method and device, electronic equipment and storage medium | |
| Ohanuba et al. | Topological data analysis via unsupervised machine learning for recognizing atmospheric river patterns on flood detection | |
| CN109885797B (en) | Relational network construction method based on multi-identity space mapping | |
| CN115982654A (en) | Node classification method and device based on self-supervision graph neural network | |
| CN112463974A (en) | Method and device for establishing knowledge graph | |
| CN107423319B (en) | Junk web page detection method | |
| CN109308565B (en) | Crowd performance grade identification method and device, storage medium and computer equipment | |
| CN113886547A (en) | Client real-time conversation switching method and device based on artificial intelligence and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |