CN105046601A

CN105046601A - User data processing method and system

Info

Publication number: CN105046601A
Application number: CN201510402086.5A
Authority: CN
Inventors: 潘星成; 荆磊; 胡俊文; 赵迪龙; 蔡秉汉
Original assignee: Culture Media (shanghai) Co Ltd
Current assignee: Culture Media (shanghai) Co Ltd
Priority date: 2015-07-09
Filing date: 2015-07-09
Publication date: 2015-11-11

Abstract

The invention provides a user data processing method and system. The method includes the following steps that: user data are acquired through a plurality of information sources; resident identification numbers corresponding to current residents are extracted from a hotel PMS; the user data acquired through the plurality of information sources respectively are analyzed real time according to the resident identification numbers, and target data corresponding to the current residents are extracted from the user data respectively; and data integration is performed on the target data, so that resident data of a unified format can be obtained. The method and system provided by the invention can support client ends of different types and can acquire user behavior data fast and conveniently. According to the method and system, user attribute data and user behavior data of the target users can be integrated, so that the user data can be generated in real time through a plurality of ways, and therefore, management of user data acquisition modes of different types of information sources can be realized, and the features of the target users can be obtained timely. With the method and system adopted, active input performed by the users who are using televisions can be avoided as much as possible, and operating efficiency and user experience can be enhanced.

Description

User data disposal route and system

Technical field

The present invention relates to data acquisition and procession, particularly, relate to user data disposal route and system.

Background technology

At current user profile gatherer process, often need user manually to input information to be registered, although the information collected like this is accurate, stable, also often add the complicated processes of user operation simultaneously.

In hotel industry, the information of general each hotel application PMS system (PropertyManagementSystem) to dweller, room and bill manages, dweller is after hotel handles information registration or buys some service in hotel, and PMS system can store personal information and the behavioral data of dweller; Hotel industry OTA refers to the online travel service commission merchant of hotel third party (OnlineTravelAgency, be called for short OTA), when user is placed an order to hotel by the website of OTA or client, the service platform of OTA also can store personal information and the subscription data of user; User moves in hotel, and after downloading the official APP using hotel, APP server also can store personal information and the information on services of user.Personal information and the behavioral data of user often come from multiple support channels, but the data layout of multiple support channels disunity, server is when analyzing the data gathering user, user is needed again manually to input personal information toward contact, as the service, hobby etc. of the cell-phone number of user, sex, preference, the process due to repeated acquisition is too loaded down with trivial details or for the consideration of privacy, user is often idle in input this type of information, what reduce information enriches degree, the statistical study efficiency of the data that also slowed down.

Summary of the invention

For defect of the prior art, the object of this invention is to provide a kind of user data disposal route and system, to solve following technical matters:

1, how automatically, in real time from user data such as information source user behavior datas;

2, the input of user how is not relied on to obtain user data;

3, the collection rule of how flexible configuration user data, thus highly versatile, can manage dissimilar user behavior data collection concentratedly.

According to a kind of user data disposal route provided by the invention, it comprises the steps:

Gathered user data respectively by multiple information source, wherein, described user data comprises UAD and user behavior data;

Dweller's identification number that current dweller is corresponding is extracted from hotel PMS system;

In real time the user data collected separately by multiple information source is analyzed according to described dweller's identification number, from described user data, extract target data corresponding to current dweller respectively;

Data Integration is carried out to described target data, obtains dweller's data of Uniform data format.

According to a kind of user data disposal system provided by the invention, comprise as lower device:

First harvester, for being gathered user data respectively by multiple information source, wherein, described user data comprises UAD and user behavior data;

First extraction element, for extracting dweller's identification number corresponding to current dweller from hotel PMS system;

Second extraction element, for analyzing the user data collected separately by multiple information source in real time according to described dweller's identification number, extracts target data corresponding to current dweller respectively from described user data;

First integrating apparatus, for carrying out Data Integration to described target data, obtains dweller's data of Uniform data format.

Compared with prior art, the present invention has following beneficial effect:

The present invention is gathered user data respectively by multiple information source, and wherein, described user data comprises UAD and user behavior data; Dweller's identification number that current dweller is corresponding is extracted from hotel PMS system; In real time the user data collected separately by multiple information source is analyzed according to described dweller's identification number, from described user data, extract target data corresponding to current dweller respectively; Data Integration is carried out to described target data, obtains dweller's data of Uniform data format.Flexibly, have the feature of highly versatile, be not limited to the advantage of specific transactions (client), what improve information enriches degree, can carry out statistical study in real time to the data that different channel obtains in the collection rule definition of user data of the present invention; By abundant user data can be obtained by all kinds of means, and then can user characteristics be understood more accurately and realize intelligent recommendation with this, and avoid user initiatively to input when using TV as far as possible, enhance operating efficiency and Consumer's Experience.

The present invention after user registers as current dweller in hotel PMS system, then extracts target data corresponding to current dweller from the user data of multiple information source; Achieve and be registered as trigger point with the dweller of hotel PMS system, complete the process of collection user data-trigger target data-integration dweller data, avoid before user becomes dweller the advanced processing that user data carries out, to alleviate data processing load, the last moment becoming dweller user gets up-to-date user data, to reflect the feature of user more accurately.

Accompanying drawing explanation

By reading the detailed description done non-limiting example with reference to the following drawings, other features, objects and advantages of the present invention will become more obvious:

Fig. 1 is the schematic flow sheet of user data disposal route provided by the invention;

Fig. 2 is the structural representation of user data disposal system provided by the invention;

Fig. 3 is the principle schematic of user data Acquire and process in the present invention;

Fig. 4 is the principle schematic to user's recommendation information in the present invention.

Embodiment

Below in conjunction with specific embodiment, the present invention is described in detail.Following examples will contribute to those skilled in the art and understand the present invention further, but not limit the present invention in any form.It should be pointed out that to those skilled in the art, without departing from the inventive concept of the premise, some changes and improvements can also be made.These all belong to protection scope of the present invention.

The invention provides one and can support dissimilar client, gather the method and system of user behavior data quickly and easily.Have employed the integrated technical scheme of client, hotel PMS system, online booking canalization in the present invention, by UAD and the user behavior data of conformity goal user, user data is generated by all kinds of means in real time, thus achieve the user data acquisition mode of the dissimilar information source of management, and then understand targeted customer's feature in time.

As shown in Figure 2, according to a kind of user data disposal system provided by the invention, comprise as lower device:

First integrating apparatus, for carrying out Data Integration to described target data, obtains dweller's data of Uniform data format;

First generating apparatus, for described dweller's data are carried out attribute extraction, generates the elementary label of guest;

First clustering apparatus, for utilizing set guest's model to carry out cluster on elementary for described guest label, obtains cluster result;

3rd treating apparatus, optimizes label for obtaining the secondary label of guest according to described cluster result as guest;

First recommendation apparatus, recommends corresponding destination service information for optimizing label according to described guest to current dweller.

Preferably, described first extraction element, comprises as lower device:

First acquisition device, for obtaining information carrier corresponding to current dweller by hotel's PMS system, comprises dweller's identification number that current dweller is corresponding in described information carrier, also include following any one or appoint much information:

-out of Memory source is corresponding except hotel PMS system identification number;

-the user behavior data that collected by out of Memory source except hotel PMS system;

-the UAD that collected by out of Memory source except hotel PMS system;

First resolver, for resolving described information carrier, obtains dweller's identification number that current dweller is corresponding.

Preferably, described second extraction element, comprises as lower device:

-the second acquisition device, for the instruction according to identification number corresponding to described out of Memory source, obtains the general character user data of colony belonging to dweller as target data from described out of Memory source; And/or

-three acquisition device, for the instruction according to identification number corresponding to described out of Memory source, accesses described out of Memory source, and obtains the target data corresponding to dweller's identification number according to the dweller's identification number being obtained from described information carrier from described out of Memory source.

Preferably, described second extraction element, comprises as lower device:

4th acquisition device, resolves respectively for the described user data gathered each information source, obtains each self-corresponding property content of each user data respectively;

3rd extraction element, for carrying out extract real-time respectively in described property content according to described dweller's identification number, using the property content associated corresponding with described dweller's identification number in described user data as target data.

Preferably, described first integrating apparatus, comprises as lower device:

First creation apparatus, for creating the Data Entry recording described dweller's identification number;

First recording device, for being recorded in attribute field corresponding to described Data Entry by extraction respectively from the target data of each user data;

First treating apparatus, for using the dweller data of described Data Entry as Uniform data format corresponding to described dweller's identification number.

Preferably, described first recording device, comprises as lower device:

First determining device, for determine described target data from the type of information source;

Second treating apparatus, for the type according to described information source, extracts the aiming field of existence from described target data, and obtains the described aiming field correlation rule corresponding with the type of information source; Wherein, described information source to be distinguished target data by described aiming field and is identified, and described correlation rule is used to indicate the corresponding relation of aiming field respectively and in Uniform data format between each attribute field;

Second recording device, for according to described correlation rule, is recorded target data corresponding for different target field respectively in attribute field corresponding in Uniform data format, forms dweller's data of described Uniform data format.

In more preference:

Preferably, described first acquisition device, comprises as lower device:

First control device, for making PMS system in hotel from an information source first information carrier, wherein, comprises dweller's identification number that current dweller is corresponding in described first information carrier, also include following any one or appoint much information:

The identification number of-described another or another multiple information source;

-the user behavior data that collected by another information source described;

-the UAD that collected by another information source described;

Second resolver, for resolving described first information carrier, obtains dweller's identification number that current dweller is corresponding;

Described second extraction element, comprises as lower device:

5th acquisition device, for according to the instruction of identification number being received from described in a described information source another or another multiple information source, obtains the general character user data of colony belonging to dweller as target data from described another or another multiple information source; And/or

6th acquisition device, for according to the instruction of identification number being received from described in a described information source another or another multiple information source, access described another or another multiple information source, and obtain the target data corresponding to dweller's identification number according to the dweller's identification number being received from a described information source from described another or another multiple information source;

A described information source is mobile phone, described first information carrier is Quick Response Code, another information source described is hotel PMS system or OTA system, and the identification number of described another or another multiple information source extracts the behavior record file produced from another or another multiple information source described in described mobile phone access;

Preferably, described first acquisition device, comprises as lower device:

First reading device, for reading the second information carrier from hotel's PMS system; Wherein, described second information carrier is generated by hotel PMS system and/or the out of Memory source except hotel PMS system, and implants dweller's identification number corresponding to current dweller, also implant following any one or appoint much information:

The identification number in-described out of Memory source;

-the user behavior data that collected by described out of Memory source;

-the UAD that collected by described out of Memory source;

3rd resolver, for resolving described second information carrier, obtains dweller's identification number that current dweller is corresponding;

Preferably, described when Data Integration is carried out to described target data:

Carry out Data Integration by information source to target data, information source comprises as lower device:

-first information source treating apparatus, for gathering user data according to Uniform data format; Or

-the second information source treating apparatus, for being converted to Uniform data format by the user data collected.

Described user data disposal system, also comprises as lower device:

First dispensing device, for sending data format instructions to information source, wherein, data format instructions is used for indicating described Uniform data format to information source.

Described data format instructions comprises collection rule, correlation rule, format information, wherein:

Described collection rule is used for the aiming field indicating the user data that need gather to information source, and wherein, information source to be distinguished user data by described aiming field and identified;

Described correlation rule is used for the corresponding relation between information source indicating target field and format information;

Described format information is used for the record position of user data in described Uniform data format to information source instruction different target field.

Described first dispensing device, comprises as lower device:

First determining device, for the type in comformed information source;

Second generating apparatus, for the type according to information source, generates the collection rule corresponding with format information and correlation rule;

Second dispensing device, for by comprise format information, collection rule, correlation rule Uniform data format send to information source.

Described Uniform data format comprises some subregions, wherein any one or appoint a multiple subregion be described below any one or appoint multiple format information:

-information source type;

-information source identification number;

Data acquisition equipment type in-information source;

-data acquisition equipment identification number;

-customer identification number;

-user's operation information;

-application interface type;

-application identities number;

-application version number;

-application interface identification number;

-application interface father interface identification number;

-content information;

-content type;

-contents attribute;

-behavior label.

Described data format instructions is also used to indicate information source to carry out burying a little at application interface according to the collection rule in Uniform data format.

Described data format instructions is also used to indicate time or the network condition that information source sends described user data.

Preferably, described multiple information source comprises hotel PMS system, also comprises subscription client and/or OTA system.

Preferably, described second extraction element, comprises as lower device:

7th acquisition device, for obtaining UAD from hotel PMS system, wherein, described UAD comprises dweller's identification number;

8th acquisition device, for obtaining corresponding user data according to dweller's identification number from another outside hotel PMS system or another multiple information source, and then extracts described target data.

Particularly, as shown in Figure 1, described user data disposal system can by realizing according to a kind of user data disposal route provided by the invention of illustrating as follows, and wherein, described user data disposal route comprises the steps:

Just the described user data gathered respectively by multiple information source is described below.

As shown in Figure 3, described multiple information source comprises hotel PMS system, also comprises client and/or web station system, and client can be the client in hotel guest room, also can be the client of user individual; Web station system can be OTA (OnlineTravelAgent is online tourism society) website.

Described user data can be corresponding data according to different information sources, specific as follows:

When information source is the PMS system in hotel, described user data comprises calendar information and/or the identity information of user, and using the identity information of user as user totem information, wherein, identity information comprise name, certificate number, contact method (such as phone number, E-mail address, micro-signal, No. QQ etc.).

When active user check-ins to hotel, the user data of the PMS system acquisition active user in hotel, as active user name, identification card number (passport No.), sex, move in the time, move in room, move in number of days, move in number, contact method, with the name, identification card number etc. of pedestrian.Handle active user and completely moved in formality, when namely representing that active user moves in, then triggered whole user data processing procedure, the televisor installed by the room moved in active user display recommendation information.

Again such as, described user data also comprises the much information in client and/or web station system: operation note, browse record, subscription information.Such as, the intelligent television client enterprising line operate of user in hotel and browsing, then can gather Operation Log corresponding to this intelligent television client and/or travel log, extract log content corresponding in described Operation Log and/or travel log, and described log content is analyzed, key word analysis extraction obtained is as user data.Particularly, user data can also be by each page in recording user is checked in the room moved in respectively intelligent television and the corresponding residence time, each menu option of triggering and the menu entries of correspondence, add up the residence time of each page and correspondence, the operating frequency of each menu entries respectively, extract content of pages under menu option in the forward page of rank and the page residence time and content, and press the weights statistics keyword of each ingredient in content of pages (title, subtitle, text), keyword statistics obtained is as user data.

Again such as, user conducts interviews and operates in OTA website, can obtain browsing information and operation information in access log, Operation Log, and extracts browsing information and operation information and obtain key word and like preference as user data with characterizing consumer.Web station system, except OTA website, can also be restaurant's Ticket booking system, ticket booking system, film seat reservation system, taxi reservation system, other PMS system etc. can obtain the system of more users information.

Again such as, can also from the internet records of user smart mobile phone, extract content of pages in APP and APP that mobile phone terminal browses and keyword thereof as user data.Preferably, hotel's client comprises the intelligent TV set in the room being arranged on Mei Jian hotel, and intelligent TV set associates with the smart mobile phone of user in advance, and smart mobile phone sending controling instruction is controlled intelligent television.When user is by smart mobile phone remote-controlled intelligent TV, the operation note of the telecommand recording user received by intelligent TV set, according to operation note, Operation Log corresponding to active user and/or travel log are extracted, extract the operation note of user when RTV remote television and browse record, the operation note corresponding in conjunction with user and browse the interest of record analysis user, using the key word corresponding with user interest as user data.Same, the statistics keyword such as each page in mobile phone A PP and the corresponding residence time, the operating frequency of each actions menu entry, the content of text of user's input, the information on services that made a reservation for by APP can be added up respectively, using keyword corresponding for each APP as user data.

User as one " American of 39 years old " have subscribed to moving in of Beijing hotel from ctrip.com, after hotel check-ins, the access log of this user of extract real-time in the OTA website comprising ctrip.com, obtain the info web such as dining room of tourism of Beijing attack strategy that this user of nearest a period of time browses in OTA website, hobby, analyzed by these info webs, extract key word.Further, the access log of this user mobile phone at particular type APP can also be extracted, as read subscription information in the predetermined module of APP, which type of hotel it moved in, the take-away of being which kind of type etc., more fully obtain the preference information of user; Its PMS information in other hotel can also be obtained, as ordered which type of service etc. before him in hotel; Again further, the ticketing information that this user carries out in sale of ticket website performing can also be extracted from ticketing website, obtain date and the place of this performance, this user can also be obtained at what date and time, to which destination get off, which humane classical etc. information of passage in transit from taxi reservation system.Further, extract its operation room TV time browse record and operation note.

The dweller's identification number just extracting current dweller corresponding from hotel PMS system is below described.

Described when extracting dweller's identification number corresponding to current dweller from hotel PMS system:

Obtain information carrier corresponding to current dweller by hotel's PMS system, in described information carrier, comprise dweller's identification number that current dweller is corresponding, also include following any one or appoint much information:

-the UAD that collected by out of Memory source except hotel PMS system;

Information carrier is resolved, obtains dweller's identification number that current dweller is corresponding.

Correspondingly, described when extracting target data corresponding to current dweller respectively from described user data:

-according to the instruction of identification number corresponding to described out of Memory source, the general character user data of colony belonging to dweller can be obtained from described out of Memory source as target data;

-also according to the instruction of identification number corresponding to described out of Memory source, described out of Memory source can be accessed, and obtain the target data corresponding to dweller's identification number according to the dweller's identification number being obtained from described information carrier from described out of Memory source.

Described information carrier can be first information carrier or the second information carrier, is described first information carrier or the second information carrier below by preference.

In a preference, described by hotel PMS system acquisition information carrier time:

Make PMS system in hotel from an information source first information carrier, wherein, in described first information carrier, comprise dweller's identification number that current dweller is corresponding, also include following any one or appoint much information:

-except hotel PMS system the identification number in out of Memory source;

-the UAD that collected by out of Memory source except hotel PMS system;

Described first information carrier is resolved, obtains dweller's identification number that current dweller is corresponding;

Can according to the instruction of identification number being received from out of Memory source outside described hotel PMS system, outside described hotel PMS system, out of Memory source obtains the general character user data of colony belonging to dweller as target data.Such as, from browser, obtain the text messages such as the link name linked in the web site name of the browsed website of user and webpage, from web site name and link name, propose keyword.Such as dweller often browses the webpage of Sichuan cuisine column in the website of " gourmet " by name, then can extract the keyword such as " cuisines ", " Sichuan cuisine ", thus will the general character " spicy " of the person sponging on an aristocrat colony of Sichuan cuisine, " fresh perfume (or spice) " data be liked as the target data of this dweller belonging to dweller.

Also can according to be received from a described information source comprise described in dweller's identification number of another or another multiple information source, thus according to the access of described dweller's identification number described another or another multiple information source, and obtain the target data corresponding to dweller's identification number according to the dweller's identification number being received from a described information source from described another or another multiple information source.Such as, when current dweller is check-ined by the PMS system in hotel, the PMS system in hotel reads current dweller at the OTA website lodging certificate that produces of timing in advance, in lodging certificate except the information comprising this order, also comprise the user name of current dweller at OTA website, then server searches the relevant information of active user from this OTA website according to this user name, comprise active user at the viewed film of this OTA website, from film ticket that this OTA website was bought etc., thus know the preference of current dweller, and similar film is recommended in the TV of guest room, and the film avoiding recommendation user viewed, namely the session obtaining respective site from hotel's PMS system is utilized, and obtain the personal behavior data of this user from corresponding website according to session.

A described information source can be mobile phone, described first information carrier can be Quick Response Code, another information source described can be hotel PMS system or OTA system, and the identification number of described another or another multiple information source extracts the behavior record file produced from another or another multiple information source described in described mobile phone access; Particularly, when user browses the information of the different channel such as OTA website, official website, hotel, can produce the behavior record file that have recorded user browsing behavior etc., behavior record file can comprise the unique identification data (domain name of such as website) that SessionID mechanism file (comprising SessionID) etc. is used to indicate user in can being.When the Quick Response Code that this user uses subscription client (such as mobile phone) to scan on as above website, (this Quick Response Code can be OTA website, official website, hotel produces with the service provider implementing method provided by the present invention), these unique identification data by Quick Response Code collection in mobile phone, and then from mobile phone, obtain the user data including this unique identification data, thus rely on the instruction of this unique identification data (link of such as domain name is pointed to) to access corresponding website, just can collect user behavior data by all kinds of means, namely the data that may correspond to the separate sources under same user carry out data analysis, wherein, the Common behavior data of colony belonging to user are preferably obtained from corresponding website.

First information carrier can also be the codings such as bar code, hybrid code, colour code, include in described first information carrier following any one or appoint much information:

The identification number of-described another or another multiple information source; Such as website domain name, link, server address etc.;

-customer identification number; The user name, login ID etc. of such as user.

In another preference, described by hotel PMS system acquisition information carrier time:

The second information carrier is read from hotel's PMS system, wherein, second information carrier can be the codings such as bar code, Quick Response Code, hybrid code, colour code, described second information carrier is generated by hotel PMS system and/or the out of Memory source except hotel PMS system, and implant dweller's identification number corresponding to current dweller, also implant following any one or appoint much information:

The identification number in-described out of Memory source;

-the user behavior data that collected by described out of Memory source;

-the UAD that collected by described out of Memory source;

Described second information carrier is resolved, obtains dweller's identification number that current dweller is corresponding;

Second information carrier can record the information (such as COOKIEID) in COOKIE mechanism file.Such as, when user implements the behavior such as consumption voucher, (being preferably recorded in COOKIE) integration using OTA, hotel to provide, the unique identification data such as user name, COOKIEID along with integration, consumption voucher collected in hotel PMS system.After server in present system obtains hotel PMS data, corresponding Users'Data Analysis can be carried out to the data of different channel.That is, can first from the unique identification data (phone number such as COOKIE mechanism file during registration of website member) that an information source obtains, then other information sources (such as QQ store) that the relevant information (with the QQ number of handset binding) that basis and unique identification data together receive is pointed to separately obtain user data, wherein, phone number, can be all documented in the QQ number of handset binding in a Quick Response Code as the second information carrier.

Analyze the user data collected separately by multiple information source according to described dweller's identification number with regard to described below, the target data extracting current dweller respectively corresponding from described user data is described.

Particularly, the described user data that each information source gathers is resolved respectively, obtain each self-corresponding property content of each user data respectively; Extract in described property content respectively according to described dweller's identification number, using the property content associated corresponding with described dweller's identification number in user data as target data;

Wherein, include in property content or mark dweller's identification number, if dweller's identification number that dweller's identification number that is that comprise in a property content or mark is corresponding with described current dweller is consistent, then think that this property content dweller identification number corresponding with described current dweller is corresponding to associate, namely dweller's identification number that current dweller is corresponding with include in property content or dweller's identification number of marking mate, if unanimously the match is successful for both, then think that dweller's identification number corresponding to current dweller is corresponding with property content and associate.

Such as, for using mobile phone as information source, the user data such as circle of friends of the micro-letter gathered is resolved, obtain the review information of numerous user and release news, then according to the cell-phone number as dweller's identification number in the review information of numerous user and find this review information submitted to by this cell-phone number in releasing news, what release news and release news (is contained this cell-phone number by property content such as review information in review information, to represent that this comment is submitted to by this cell-phone number), thus using the target data of these property content as this dweller, after keyword extraction and process, the hobby and focus that obtain this dweller can be analyzed.

Can also resolve from the user data of making a reservation in website be associated with above-mentioned cell-phone number, (menu name can mark this cell-phone number by property content such as menu names predetermined in this cell-phone number in history period recently to obtain dweller, to represent that this vegetable is that this cell-phone number is predetermined), thus using the target data of these property content as this dweller, the taste preference of dweller by analysis, can be obtained.

Described when extracting target data corresponding to current dweller respectively from described user data:

-from hotel PMS system, obtain UAD, wherein, described UAD comprises dweller's identification number;

-obtain corresponding user data according to dweller's identification number from another outside hotel PMS system or another multiple information source, and then extract described target data.

In a preference, described dweller's identification number is the user totem information of indicating user individual, after obtaining user totem information from hotel PMS system, user totem information (such as passport number) can be passed through and uniquely can determine that user represents the user identity that information is corresponding, thus find corresponding user data according to this user identity in each out of Memory source.Such as whether correspond to same user according to the user data in the cell-phone number identification PMS end of user, client, hotel reservation channel, if correspond to same user, then can carry out analytic statistics to the user data from each information source under same user respectively, thus the individual private information such as name, mobile phone not needing active user additionally to input oneself in other information sources directly can obtain target data corresponding to active user from other information sources.

Such as, other information source above-mentioned comprises OTA website, described OTA website when user operation recording user behavior as user data; Wherein, user behavior comprises the access log of user in OTA website, Operation Log, and these user data obtain this user interest preference by contributing to analyzing in the subsequent step of method provided by the present invention.

Data Integration is carried out to described target data be described with regard to described below.

Integrated data processing can be carried out to the user data being received from the information sources such as client, PMS Hospitality management system, other hotel reservation canalization to make the present invention, one aspect of the present invention is preferably real-time from multiple information source user data, the present invention on the other hand, Uniform data format is carried out by the user each information source obtained, the advantage that the user behavior data that can realize having management different types of information source gathers, and then understand targeted customer's feature all sidedly in time and realize intelligent content recommendation with this.

The sequencing target data got sent according to each information source correspondence or other order (such as random sequence or the prioritization according to the information source preset) are combined as a Frame as integration means, also target data can be filled in dweller's data of Uniform data format in corresponding attribute field respectively according to the data frame format preset.

It is described when Data Integration is carried out to described target data, the Uniform data format of user data is completed by data source, for this reason, information source can gather user data respectively according to Uniform data format, and the user data collected also can be converted to Uniform data format by information source respectively.

In order to enable dweller's data of Uniform data format mate with each information source, described respectively user data to be gathered by multiple information source before also comprise the steps:

Send data format instructions to information source, wherein, data format instructions is used for indicating described Uniform data format to information source.Like this, just can for the type of different information sources (such as mobile phone or TV, APP such as different again) send corresponding data format instructions respectively, make the user data of different information sources can be converted to dweller's data of Uniform data format, these user data can both be used and unified process.Wherein, the aiming field between the information source of same type is identical, can think the information source belonging to same type between such as, different editions with a APP.Certainly, if generated corresponding data format instructions for the information source of certain type before, then this aiming field instruction can be kept in dictionary table, when running into the information source of same type so that follow-up, extracting directly uses.

Described data format instructions comprises collection rule, correlation rule, attribute field, wherein:

Described correlation rule is for the corresponding relation in the dweller's data to information source indicating target field and Uniform data format between each attribute field;

Wherein, attribute field is used for the record position of user data in described Uniform data format to information source instruction different target field.

Such as, in a website as information source, read the user journal of browser server, the click behavior of user is divided into the aiming field of " super chain is clicked " and the aiming field of " choosing click ", and in Uniform data format, only there is the attribute field of one " mouse clicks ", then by analyzing the user journal of this browser server in advance, the dictionary table for this browser can be set up, the aiming field in browser server daily record is described in this dictionary table, and containing the correlation rule (i.e. corresponding relation) between the every attribute field in aiming field and Uniform data format, thus according to this dictionary table, can learn that the aiming field of " super chain is clicked " and the aiming field of " choosing click " are all associated with the attribute field of " mouse clicks " in dweller's data of Uniform data format, thus the content of text that website name and " choosing click " that " super chain is clicked " obtains obtain is recorded in attribute field corresponding to " mouse clicks ", form dweller's data of Uniform data format.

Below just described in obtain Uniform data format dweller's data be described.

Described when obtaining dweller's data of Uniform data format:

Create the Data Entry recording dweller's identification number;

Extraction is recorded in attribute field corresponding to Data Entry from the target data of each user data respectively;

Using the dweller data of Data Entry as Uniform data format corresponding to described dweller's identification number.

Described will extract when the target data of each user data is recorded in attribute field corresponding to Data Entry respectively:

Determine described target data from the type of information source;

According to the type of described information source, from described target data, extract the aiming field of existence, and obtain the described aiming field correlation rule corresponding with the type of information source; Wherein, described information source to be distinguished target data by described aiming field and is identified, and described correlation rule is used to indicate the corresponding relation of aiming field respectively and in Uniform data format between each attribute field;

According to described correlation rule, target data corresponding for different target field is recorded respectively in attribute field corresponding in Uniform data format, forms dweller's data of described Uniform data format.

In dweller's data of described Uniform data format, comprise some subregions, wherein any one or appoint a multiple subregion be described below any one or appoint multiple format information:

-information source type; Such as may be used for indication information source is smart mobile phone, intelligent television, online tourism website, APP etc.;

-information source identification number; Such as may be used for the numbering of intelligent television in the identifier of record mobile phone, hotel guest room;

Data acquisition equipment type in-information source; Such as information source is hotel PMS system, then data acquisition equipment type can be display screen, elevator keyboard, air-conditioning etc.;

-data acquisition equipment identification number; Such as may be used for the numbering recording air-conditioning;

-customer identification number; It can be such as the login name etc. of the name of user, micro-signal, game center;

-user's operation information; Can be such as the mouse number of clicks of user, the length of stay, touch-control, Sound control, gravity sensing etc. at Webpage;

-application interface type; Can be such as the map page, original list, details page etc.;

-application identities number; It can be such as APP numbering in APP store etc.;

-application version number;

-application interface identification number; It can be such as the numbering of APP median surface;

-application interface father interface identification number;

-content information; Such as information described in content etc.;

-content type; Type belonging to content information can be such as picture, text, video, link etc.;

-contents attribute; Attribute classification belonging to content information can be such as sight spot, cuisines, tradition performance etc.;

-behavior label.

Correspondingly, can encode according to unified coding rule to the user data of different-format, obtain the user data of Uniform data format.Such as, following coding rule can be adopted:

The coding in described A district, main inclusion region, equipment, subscriber-coded, for marking hotel and user.Specifically consist of: regional code | device type | device coding | user | the running time; When a multiple field of coding forms, identified by English CSV;

Mark illustrates:

-regional code: refer to encode in hotel, uniquely can identify the coding in hotel, as fishing platform; Length 4Byte;

-device type: the source referring to image data, for distinguishing TV end, mobile phone terminal, PMS system, other hotel reservation channel; Length 1Byte;

-device coding: refer to television devices unique encodings; If character appears in encoded radio, binary digit mode character being changed into ASCII character is needed to identify; Length: 30Byte;

-subscriber-coded: refer to user ID; Support that multiple user ID combines, identified, as ID1, ID2 by English CSV between multiple combination; Be no more than 10 users; Length: 40Byte;

-the running time: the concrete time referring to user operation, be accurate to Millisecond; Length: 8Byte;

Described B district coding, mainly comprises client application and is correlated with, for marking the operation interface of client.Concrete composition: application coding | interfacing code | parent page is encoded | interface type | version number encodes;

Mark illustrates:

-application coding: refer to certain client application concrete; Length: 2Byte;

-interfacing code: certain interface referring to certain client application; Length: 1Byte;

-father interfacing code: the father's interfacing code referring to current interface; Length: 1Byte;

-interface type: the type referring to interface, as map, list, details page etc.; Length: 1Byte;

-version number encodes: the concrete version number coding referring to current application; Length: 1Byte;

Described C district coding, main content is correlated with, for marking the service content that client provides.Concrete composition: research content | content type is encoded | and contents attribute is encoded;

Mark illustrates:

-research content: particular content ID, as certain cuisines content, certain inspiration video, certain circuit, certain direct broadcast band; Length: 4Byte;

-content type is encoded: content generic ID combines, as ID1, ID2, ID3; Classification refers to cuisines, KTV, SPA, market etc.; Length: 40Byte;

-contents attribute is encoded: the combinations of attributes that content is concrete, as DATA1, DATA2, DATA3; Length: 120Byte; Particularly, contents attribute refers to the details that content type is corresponding, the contents attribute of such as cuisines content type can be " home cooking ", " feast dish " etc., and the contents attribute of such as SPA content type can be " safe formula hydrotherapy resistance to compression ", " Chinese style herbal medicine toxin expelling ", " physiotherapy of French spring " etc. again.

Described D district coding, main finger behavior label, for marking the user behavior of current dweller;

Label illustrates:

-behavior label: the user behavior tag combination on time shaft, as ACT1, ACT2, ACT3; Label is as lunch, food, SPA etc.; 40Byte; The behavior tag characterization feature of user behavior;

Described E district coding, mainly refers to remote controller key, is used to indicate the operation of user.Remote controller key is arranged on mobile phone terminal by APP, produces corresponding trigger pip according to user to the operation of mobile phone terminal, the coding write E district coding that trigger pip correspondence produces;

Label illustrates:

-keyboard-coding: remote controller key; Length: 1Byte; Such as Audio mode key etc.

Described F district coding, for reserved expansion.

Further, with the message transmitting procedure of information source, due to the private information of user may be related to, AES encryption algorithm for encryption user data can therefore be adopted.And adopt HTTPPOST method transmitting user data between information source, once report all data; If report successfully, client deletes the user behavior data reported, if report failure, retains data and treats that next time reports again.

Further, described data format instructions is also used to indicate time or the network condition that information source sends described user data.The timing of the information source such as intelligent television, air-conditioning in guest room is such as indicated to provide user data, again such as, indicate by the smart mobile phone self monitor gas radio network conditions of user, report of user data under connection WIFI state, advantage is the network traffics can saving user oneself.Run into abnormal shutdown sudden strain of a muscle to move back etc. and cause loss of data, can ignore.

Further, described data format instructions is also used to indicate information source and carries out burying a little at application interface according to the collection rule in Uniform data format.Under interface concrete in the client, under concrete operating function, imbed collection point according to user behavior data collection rule, when this function of user's concrete operations, the behavioral data of user can be obtained.Such as, when gathering user data from client, gather the touch information of user, as user click or the time interval of slip button, the time interval of operation, the residence time etc. of the page, judge the hobby of user to corresponding content of pages further:

Just how to be described to dweller's recommendation service information according to dweller's data below.

Preferably, described user data disposal route, also comprises the steps:

Dweller's data are carried out attribute extraction, generates the elementary label of guest;

Utilize set guest's model to carry out cluster on elementary for guest label, obtain cluster result;

Obtain the secondary label of guest according to cluster result and optimize label as guest;

Optimize label according to guest and recommend corresponding destination service information to current dweller.

Wherein, described guest's model can be carry out attribute extraction (as adopted the statistical method of TF/IDF to historical data) to a large number of users data being converted to Uniform data format (large data) and train, set up guest's model, respectively marking cluster is carried out to each dweller's data by guest's model, namely by the method for static classification, by Data classification in the different classes of feature pre-set, thus judge certain classification tendency of dweller's data.Attribute extraction is carried out to the current data collected, mating extracting the attribute information that obtains in a model, obtaining the result of cluster, be i.e. all kinds of labels of dweller's hobby, follow-uply will recommend corresponding information on services according to these labels to dweller.As one " American of 39 years old " subscribes moving in of Beijing hotel from ctrip.com, after hotel check-ins, its user profile registered in website of server extract real-time, its cell-phone number, the serviced device of the essential informations such as age obtains, this user can receive the information on services in this hotel at any time by mobile phone, simultaneously his intelligent television also corresponding customized information of sending of reception server showing of moving in room, fill message is carried out separately without user, registration, the operation that certification etc. are loaded down with trivial details, improve the degree of enriching of information and the statistical study efficiency of data.Simultaneously, the user profile of this user extracted is mated in user model in correspondence, the corresponding score value of label such as " middle age ", " America and Europe ", " high heat food ", " the good dining room of environment ", " OTA ", " San Litun " is higher in a model, and " cross-talk ", " zoo ", " children's paradise ", " parent-offspring " then corresponding score value are on the low side.

Particularly, those skilled in the art can utilize prior art to obtain described guest's model, such as openly obtain from the third party's network managing large data service, again such as can be as shown in Figure 4, respectively by multiple user data such as client, PMS system, reservation systems in the whole network, the user data collected in history is carried out the extraction of the information such as attribute, behavior, feature, train after statistics, obtain described guest's model.Preferably, described guest's model comprises multiple classification, and wherein, each classification comprises multiple set label, and these classification and set label can be thought default by experience and obtain, and also can be obtained by sample training.

Further, describedly to comprise the steps: according to the elementary label of dweller's data genaration guest

Analysis guest's attribute information and/or guest's behavioural information obtain guest's feature;

Using any one in guest's attribute information, guest's behavioural information and guest's feature or appoint multiple information as the elementary label of guest.

When cluster is carried out to each guest elementary label, perform following steps:

By the classification of elementary for this guest tag match in guest's model;

Determine the degree of association of this classification in the elementary label of this guest and guest's model, belong to the probability of this classification as this elementary label, be designated as probability A;

A level label of guest corresponding to using set for each in this classification label as the elementary label of this guest; Wherein, in this classification, each set tag match is designated as probability B in the probability of this classification;

Determine the weighted value of each secondary label, wherein, described weighted value equals the product of probability A and probability B;

Using secondary label corresponding for some weighted values forward in descending for weighted value sequence as the information in cluster result.

The present invention carries out the extraction of guest's attribute information and guest's behavioural information to user data, thus obtain guest's feature, by guest's attribute information, guest's behavioural information, guest's feature is as the elementary label of this guest, the elementary label of a such as guest is " male sex ", " number of visits of tourism natural scenic spot webpage is more ", " have purchased charge garment ", " backpacker ", wherein, elementary label " male sex " is the customer attribute information from hotel PMS system, elementary label " number of visits of tourism natural scenic spot webpage is more ", " have purchased charge garment " and be the user behavior information from guest's mobile phone browser, " backpacker " guest's feature for going out according to customer attribute information and user behavior information extraction.

Then each elementary label each given categorisation respectively in set guest's model of guest is mated, judge each elementary label respectively with each degree of association of classifying, thus cluster marking is carried out in guest's model, obtain the probability that described elementary label belongs to each classification in guest's model, and using the classification of maximum probability as the classification with corresponding elementary tag match.Further, then extract set label from classification, the guest obtaining this dweller optimizes label.

Such as, the attribute information of the current dweller of extraction and elementary label corresponding to behavioural information are " 40 years old ", " people ", " American " (extracting in PMS system), " OTA reservation ", " oxtail soup ", " century Tan Fu " (extraction of OTA system), " take-away wine ", " take-away Pizza " (cell-phone customer terminal extraction), obtaining elementary label after mating in guest's model with matching separately classify and the situation of match-percentage is " middle aged-95% ", " American-European-90% ", " high heat food-80% ", " the good dining room-60% of environment ", " network intelligent-90% ", " zoo-5% ", " cross-talk-10% ", secondary label " middle high price " corresponding in classifying in " middle age ", " SPA ", secondary label " love wine " set in " America and Europe " classification, " Christian religion ", " western-style food ", secondary label " Pizza " set in " high heat food " classification, " fried chicken ", " cola ", set secondary label " San Litun " corresponding in " the good dining room of environment " classification, " leopard ", the set secondary label " wifi " in " network intelligent " classification, " on-line payment " combines corresponding match-percentage respectively and mates, such as, in the present embodiment, in being separated with this respectively for the match-percentage of classifying by elementary label, each secondary label each self-corresponding probability score in this classification is multiplied, obtain the score value of each secondary label, then by " SPA ", " San Litun ", " Christian religion ", the secondary label that " western-style food " decile value is higher optimizes label as guest.

In a preferred embodiment, by multiple information sources such as the websites such as smart mobile phone, ticket reservation website, hotel PMS systems, user data is gathered, now data acquisition for user might not for the dweller relevant to hotel, but the public users having become or may become future dweller relates to the wider data in face screen in follow-up flow process to obtain.Namely before a user registers as dweller, the user data of this user is in unactivated state, only after user registers as dweller, just activates and obtains the user data of this user and to go forward side by side row relax.That is, hotel's PMS system is the identification point that user becomes dweller, and before user becomes dweller, the behavioral data of user etc. all belong to described user data, instead of described dweller's data or target data; After multiple information source gathers user data, and just become dweller's data of this user at user's user data that this dweller is corresponding after hotel's registration.Certainly, user herein also can Shi Yu hotel the personnel that are correlated with, the registered members in such as hotel.

In numerous users, illustrate for certain user and how to carry out user data collection the user data gatherer process of all the other users (can with reference to the user data gatherer process of this user), smart mobile phone collects the sex man of this user (this user will be called the dweller in hotel follow-up) by meter step APP, the UAD at age 30 and the user behavior data of 5 hours deep sleep time every day, ticket reservation website collects this user by the navigation patterns of this user and likes the user behavior data of viewing NBA ball match and swimming championship and this user to have purchased the clubbable UAD in fitness center, comfortable fort with the form of Quick Response Code.When this user is after hotel registers as dweller, the individual email address of this dweller can be obtained as dweller's identification number from hotel PMS system.Then exercise data corresponding to this individual email address is collected according to this individual email address from the background server of meter step APP, also collect this individual NBA of email address institute registered user and history ticketing information of swimming contest according to this individual email address from ticket reservation website, Quick Response Code can also be scanned obtain fitness center, comfortable fort membership information from reward voucher, wherein, the process analyzed user data can be dweller's identification number and the process of mating between user data, extracts target data after the match is successful from corresponding user data.

Further, the target data of this dweller is integrated into a Frame, wherein, this Frame describes the age information of this dweller obtained from multiple information source by different field, gender information, sleep info, ball match view and emulate the target information such as information, body-building membership information; Again further, by being trained the guest's model obtained by large data, obtain this dweller and belong to the sports enthusiast that the length of one's sleep is few, like motion, high consumption, therefore by hotel guest room TV to its recommend hotel nearest provide the address information of the fitness center of swimming pool as destination service information, and recommend the quietly tourist attractions such as museum unlike it.Wherein, reward voucher can be the reward voucher that hotel and fitness center, comfortable fort are combined the integration that provides and shared, Quick Response Code on it have recorded hotel information and fitness center, comfortable fort membership information, thus PMS system in hotel can obtain fitness center, the comfortable fort membership information of this dweller by scanning Quick Response Code.

In a change case, this Quick Response Code also can be shown by the smart mobile phone of dweller; In another change case, also can Shi Youyu hotel cooperation fitness center, comfortable fort Quick Response Code is attached to public place, dweller scanned this Quick Response Code by smart mobile phone before becoming dweller, the domain name of website, comfortable fort is recorded in this Quick Response Code, thus the membership information registered when this dweller browses comfortable fort health and fitness site can be recorded in smart mobile phone, thus also can obtain fitness center, comfortable fort membership information as destination service information from the smart mobile phone of dweller.

Further, described multiple information source, comprise the information source that user is known before hotel registers as dweller, the side information source found according to the instruction of dweller's identification number after obtaining dweller's identification number can also be included in, such as, before user registers as dweller, only by its microblogging, micro-letter circle of friends is as multiple Given information source, to be handled by PMS system active user and checked in, after becoming dweller, PMS system in hotel also obtains the individual email address of current dweller, thus have found current dweller by the ticketing website of this individual email address registration and social networks information source as a supplement according to individual email address as dweller's identification number, then from described Given information source and described side information source, obtain the user data that comprehensively this dweller is corresponding.

Those skilled in the art will know that, except realizing except system provided by the invention and each device thereof in pure computer readable program code mode, system provided by the invention and each device thereof can be made to realize identical function with the form of logic gate, switch, special IC, programmable logic controller (PLC) and embedded microcontroller etc. by method step being carried out programming in logic completely.So system provided by the invention and every device thereof can be considered to a kind of hardware component, and to the structure that also can be considered as the device realizing various function in hardware component comprised in it; Also the device being used for realizing various function can be considered as not only can be implementation method software module but also can be structure in hardware component.

Above specific embodiments of the invention are described.It is to be appreciated that the present invention is not limited to above-mentioned particular implementation, those skilled in the art can make a variety of changes within the scope of the claims or revise, and this does not affect flesh and blood of the present invention.When not conflicting, the feature in the embodiment of the application and embodiment can combine arbitrarily mutually.

Claims

1. a user data disposal route, is characterized in that, comprises the steps:

2. user data disposal route according to claim 1, is characterized in that, described when extracting dweller's identification number corresponding to current dweller from hotel PMS system:

-the UAD that collected by out of Memory source except hotel PMS system;

Described information carrier is resolved, obtains dweller's identification number that current dweller is corresponding.

3. user data disposal route according to claim 2, is characterized in that, described when extracting target data corresponding to current dweller respectively from described user data:

-according to the instruction of identification number corresponding to described out of Memory source, obtain the general character user data of colony belonging to dweller from described out of Memory source as target data; And/or

-according to the instruction of identification number corresponding to described out of Memory source, access described out of Memory source, and obtain the target data corresponding to dweller's identification number according to the dweller's identification number being obtained from described information carrier from described out of Memory source.

4. user data disposal route according to claim 1, is characterized in that, described when analyzing the user data collected separately by multiple information source in real time according to described dweller's identification number:

The described user data that each information source gathers is resolved respectively, obtains each self-corresponding property content of each user data respectively;

In described property content, extract real-time is carried out respectively, using the property content associated corresponding with described dweller's identification number in described user data as target data according to described dweller's identification number.

5. user data disposal route according to claim 1, is characterized in that, described in when obtaining dweller's data of Uniform data format:

Create the Data Entry recording described dweller's identification number;

Extraction is recorded in attribute field corresponding to described Data Entry from the target data of each user data respectively;

Using the dweller data of described Data Entry as Uniform data format corresponding to described dweller's identification number.

6. user data disposal route according to claim 5, is characterized in that, described will extract when the target data of each user data is recorded in attribute field corresponding to described Data Entry respectively:

Determine described target data from the type of information source;

7. user data disposal route according to claim 1, is characterized in that, also comprise:

Described dweller's data are carried out attribute extraction, generates the elementary label of guest;

Utilize set guest's model to carry out cluster on elementary for described guest label, obtain cluster result;

Obtain the secondary label of guest according to described cluster result and optimize label as guest;

Optimize label according to described guest and recommend corresponding destination service information to current dweller.

8. a user data disposal system, is characterized in that, comprises as lower device:

9. user data disposal system according to claim 8, is characterized in that, described first extraction element, comprises as lower device:

-the UAD that collected by out of Memory source except hotel PMS system;

10. user data disposal system according to claim 9, is characterized in that, described second extraction element, comprises as lower device:

11. user data disposal systems according to claim 8, is characterized in that, described second extraction element, comprises as lower device:

12. user data disposal systems according to claim 8, is characterized in that, described first integrating apparatus, comprises as lower device:

13. user data disposal systems according to claim 8, is characterized in that, also comprise: