CN105447708A - Information offering method and information offering device - Google Patents
Information offering method and information offering device Download PDFInfo
- Publication number
- CN105447708A CN105447708A CN201410433045.8A CN201410433045A CN105447708A CN 105447708 A CN105447708 A CN 105447708A CN 201410433045 A CN201410433045 A CN 201410433045A CN 105447708 A CN105447708 A CN 105447708A
- Authority
- CN
- China
- Prior art keywords
- commodity sign
- user
- identified
- time
- commodity
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an information offering method and an information offering device in order to improve the accuracy of recommendation information offered to users. The information offering method comprises the steps of acquiring an image collected by a user, identifying the content information in the image, determining the recommendation information to be offered to the user according to the content information, and offering the recommendation information to the user. According to the invention, the recommendation information to be offered to the user is determined according to the content information in the image provided by the user, the recommendation information is offered to the user, and recommendation information is not offered to the user blindly. Therefore, the accuracy of recommendation information offered to users can be improved effectively.
Description
Technical field
The application relates to field of computer technology, particularly relates to a kind of information providing method and device.
Background technology
Along with the development of internet, increasing user selects to enter to search for the information wanted on the net, in order to make user fast obtain the information wanted accurately, at present, has occurred by active for user provides the information providing technique of information.
Information providing technique is the behavioural habits according to user, promptly and accurately provide relevant information to user.Current information providing method mainly comprises: server obtain user on-line operation behavior on the internet (namely as, described shopping or search operation behavior) on-line operation record, the behavioural habits of user are determined according to on-line operation record, determine according to behavioural habits the relevant information being applicable to this user, this relevant information is supplied to user.
Concrete, for a certain user, above-mentioned information providing method needs this user often to carry out on-line operation, according to a large amount of on-line operation records of this user, determines behavioural habits, and then determine relevant information, but if this user infrequently carries out on-line operation, above-mentioned information providing method just cannot obtain a large amount of on-line operation records of this user, also just cannot determine the behavioural habits of this user, and then also just cannot provide for this user the relevant information being applicable to this user.If blindly provide information for this user, then can because of the information provided the inaccurate problem causing the wasting of resources.
Summary of the invention
The embodiment of the present application provides a kind of information providing method and device, in order to rise to the accuracy of the information that user provides.
Information providing method described in the application comprises:
Obtain the image that user gathers;
Identify the content information in described image;
Determine to be supplied to the recommendation information of user according to described content information;
Described recommendation information is supplied to described user.
Information provider unit described in the application comprises:
Image collection module, for obtaining the image that user gathers;
Content information identification module, for identifying the content information in described image;
Recommendation information determination module, to be supplied to the recommendation information of user for determining according to described content information;
Recommendation information provides module, for described recommendation information is supplied to described user.
The information providing method that the application provides, comprising: obtain the image that user gathers; Identify the content information in described image; Determine to be supplied to the recommendation information of user according to described content information; Described recommendation information is supplied to described user.The application for recommendation information that user provides be that content information in the image gathered according to user is determined, instead of to determine according to a large amount of on-line operation records of this user, therefore, when user infrequently carries out on-line operation, by obtaining the content information in the image of user's collection, determine the behavioural habits of user, and then be defined as the recommendation information that user provides, thus effectively can rise to the accuracy of the recommendation information that user provides.
Accompanying drawing explanation
Accompanying drawing described herein is used to provide further understanding of the present application, and form a application's part, the schematic description and description of the application, for explaining the application, does not form the improper restriction to the application.In the accompanying drawings:
One of information providing method schematic flow sheet that Fig. 1 provides for the embodiment of the present application;
The information providing method schematic flow sheet two that Fig. 2 provides for the embodiment of the present application;
The information provider unit structural representation that Fig. 3 provides for the embodiment of the present application.
Embodiment
In the prior art, for the user often carrying out on-line operation, a large amount of on-line operation records at general family can determine the recommendation information that will provide for user, but, in practical application scene, not necessarily each user is frequently by on-line operation behavior obtaining information, and for the user infrequently carrying out on-line operation, prior art will cannot provide recommendation information accurately for user because obtaining the on-line operation record of user.And in the embodiment of the present application, the content information in the image that the network equipment can provide according to user, is defined as the recommendation information that user provides, and be supplied to user.Because the content information in image does not rely on on-line operation, therefore, for the user infrequently carrying out on-line operation, according to the content information in the image that user provides, accurately can determine the behavioural habits of user, thus can accurately determine to be supplied to the recommendation information of user, and then effectively can improve the accuracy of recommendation information.
For making the object of the application, technical scheme and advantage clearly, below with reference to the application's specific embodiment and corresponding accompanying drawing, technical scheme is clearly and completely described.Obviously, described embodiment is only some embodiments of the present application, instead of whole embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not making the every other embodiment obtained under creative work prerequisite, all belong to the scope of the application's protection.
The information providing method will illustrating the embodiment of the present application and provide below.
See Fig. 1, be the schematic flow sheet of the information providing method that the embodiment of the present application provides, the method comprises:
The image that S101, acquisition user gather.
In the embodiment of the present application, user by the image acquisition device image in its terminal, and uploads to the network equipment (e.g., server), and the network equipment then can obtain the image that user gathers.
Concrete; in practical application scene; user often will pass through the image uploading of image acquisition device to server by social client (social APP); and be shared with other people; such as, people can be taken pictures to oneself more interested object by mobile phone usually whenever and wherever possible, and are uploaded onto the server by the social APP on mobile phone by the image obtained of taking pictures; to be shared with good friend, server then can obtain the image that user uploads.
S102, the content information identified in described image.
Wherein, image described in the embodiment of the present application can be all types of images based on different images content, as landscape image, ambient image, food image, bill images etc., accordingly, content information in image can be the content information of the graph style such as building, food in landscape, environment, and/or the content information of the literal type (literal type can comprise Chinese character type and numeric type etc.) such as landscape title, building title, food name, shopping information in bill.Such as, the content information in image can be park name, recreation ground title, eating and drinking establishment's trade name, food name, or the word such as consumption information in bill, or can be the figure of a certain landmark building in environment, as the figure of Eiffel Tower.
Be ambient image with image below, and the content information in image comprise park name is example explanation.
Suppose that user acquires by mobile phone the photo that comprises park name on doorway, vast park, and uploaded onto the server by the social APP on mobile phone, server, after getting the photo that user uploads, identifies in this photo the content information comprising " vast park " this Chinese character type by image recognition technology.
S103, determine according to described content information to be supplied to the recommendation information of user.
Concrete, the relevant information corresponding with this content information preset, after identifying the content information in image, if comprise the current content information identified in content information of determining to prestore, is then defined as to be supplied to the recommendation information of user by server.
Adopt above-mentioned example, server is after identifying in photo the content information comprising " vast park " this Chinese character type, determine to prestore in content information and comprise " vast park " this content information, the default relevant information corresponding with " vast park " is defined as to be supplied to the recommendation information of user.
Wherein, corresponding with " vast park " default relevant information can be all kinds of performing active information, admission ticket favor information etc. recent in vast park.
S104, described recommendation information is supplied to described user.
Adopt above-mentioned example, server determine to be supplied give user recommendation information after, at least one in all kinds of performing active information recent in vast park, admission ticket favor information can be supplied to user.
The information providing method as shown in Figure 1 that the embodiment of the present application provides, even if user infrequently carries out the operation of on-line search information, namely the operation note of a large amount of on-line search information is not provided to server, the image that also can upload according to user of server provides the recommendation information for this image for this user, thus effectively can rise to the accuracy of the recommendation information that user provides.
In the step S102 shown in Fig. 1, when the content information in the image that server gets is the content information of literal type, the content information of literal type in following three kinds of embodiment recognition images can be adopted.
Embodiment one: the word in recognition image, and determine the participle that is made up of the word identified, when comprising predetermined keyword in the participle determined, this participle is defined as the content information identified.
Adopt above-mentioned example, suppose that predetermined keyword comprises " park ", word then in server recognition image, and determine to comprise predetermined keyword " park " in the participle " vast park " be made up of the word identified, therefore " vast park " is defined as the content information identified.
Embodiment two: the word in recognition image, and determine the participle that is made up of the word identified, when comprising predetermined keyword in the participle determined, and when the word be positioned at after the participle determined is the word of specified type, the word of this specified type is defined as the content information identified.
Such as, suppose that user has taken a photo, comprise the content information of eating and drinking establishment's trade name in photo, if trade name is " Qianmen Peking Duck Restaurant 302 ", this trade name represents: branch sequence number is 302 and address, branch certain Peking Duck Restaurant at Qianmen.Suppose to need the content information identified to be branch sequence number " 302 " this content information, the mode of identification serial number is as follows:
Set predetermined keyword as " Peking Duck Restaurant ", and the word of setting specified type is the word of numeric type, word then in server recognition image, and determine to comprise predetermined keyword " Peking Duck Restaurant " in the participle " Qianmen Peking Duck Restaurant " be made up of the word identified, and the word be positioned at after participle " Qianmen Peking Duck Restaurant " is the word 302 of numeric type, the word " 302 " of this numeric type is defined as the content information identified.
Embodiment three: the word in recognition image, and determine the participle that is made up of the word identified, when the word that the word identified is specified type, and when determining that the participle be made up of the word of this specified type falls into default set, the participle that the word of this specified type is formed is defined as the content information identified.
In practical application scene, the content information in some image may comprise the content information of date type.Suppose in image, to comprise " on 05 28th, 2014 " this content information, identify " on 05 28th, 2014 " this content information, specific as follows:
Such as, the word of setting specified type is the word of numeric type, and the default set of setting comprises: the 3rd set of 31 participles of 01 to 31 is gathered, comprised in first set of 15 participles of 2000 to 2014, second of 12 participles of 01 to 12.
The word of the numeric type then in server recognition image, and determine that the participle be made up of the word of the numeric type identified is: " 2014 ", " 05 " and " 28 ", and determine that participle " 2014 " falls into the first set, participle " 05 " falls into the second set, participle " 28 " falls into the 3rd set, therefore, server then determines that these three participles are the content information identified, and this content information is expression 2014 05 month 28 these date and time informations.
Identify in the method for the content information of literal type at above-mentioned three kinds, server is all by optical character identification (OpticalCharacterRecognition, OCR) word in technology recognition image, and determine by the dictionary preset or participle model the participle that is made up of the word identified.
In addition, in step s 102, when the content information in the image that server gets is the content information of graph style, the content information of graph style in following methods recognition image can be adopted.
Server can determine the similarity of the image that gets and the image prestored, when the similarity determined is greater than setting threshold value, then direct using this image of prestoring content information as the graph style identified.
Further, due to compared to landscape image, the image of the comparatively changeable complexity of the colors such as ambient image, it is single that bill images has color, prospect and the clearly demarcated feature of background, therefore, from bill images, identify that the content information of literal type has higher accuracy, and, increasing user starts employing and the mode that the bill images of collection is uploaded onto the server is kept accounts, like this, server also can get corresponding bill images, therefore, be bill images by the image got for server below, describe server provides information to user method according to bill images in detail, as shown in Figure 2.
The process that the server that Fig. 2 provides for the embodiment of the present application provides information according to the bill images got to user, specifically comprises the following steps:
S201: server obtains the bill images that user gathers.
In practical application scene, user is after solid shop/brick and mortar store buys commodity, and by the corresponding bill images of terminal collection, then uploaded onto the server by this bill images by the book keeping operation APP that terminal is installed, server then gets this bill images.
Wherein, in the embodiment of the present application, described bill images take bill as the image that object gathers, dissimilar according to bill, and this bill can be, but not limited to the bill into various type of credentials, e.g., and invoice, document, shopping receipt etc.According to the different content of bill, bill can be, but not limited to for the relevant bill of staying, with relevant bill of having dinner, the bill relevant with tourism, the bill relevant with Commodity imports and exports, or to relevant bill etc. of doing shopping.
S202: the content information of literal type in server identification bill images.
In the embodiment of the present application, described content information can comprise commodity sign in bill images, goods amount and at least one in the time buying.
Be shopping receipt below with bill type, bill images is the image of shopping receipt is example explanation.
Such as, suppose that user is taken pictures to a shopping receipt by mobile phone, the image uploading obtained taking pictures is to server, server is then after the image being got this shopping receipt by step S201, by image recognition technology, image is identified, obtain commodity sign, goods amount and one or more content informations in the time buying of comprising in the image of shopping receipt.
Suppose the commodity sign comprising four kinds of commodity in shopping receipt, be respectively skirt, swimming suit, toothpaste and slippers, the goods amount that these four kinds of commodity are corresponding is respectively P1, P2, P3, P4, and the time buying that each commodity are corresponding is t1.After then server obtains the image of this shopping receipt by above-mentioned steps S201, by step S202 identify commodity sign in this image (as, skirt, swimming suit, toothpaste and slippers), goods amount (as, goods amount P1, P2, P3, P4 that each commodity are corresponding), with at least one in the time buying (the time buying t1 that e.g., each commodity sign is corresponding).
S203: server, according to the content information in the bill images identified, is determined to be supplied to the recommendation information of user.
Concrete, server is after the content information identifying literal type in bill images, comprise the current content information identified if determine to prestore in content information, then the relevant information corresponding with this content information is defined as to be supplied to the recommendation information of user.
Continue to adopt above-mentioned example, after server identifies commodity sign skirt, swimming suit, toothpaste and slippers, if determine to comprise skirt prestoring in commodity sign, then the merchandise news corresponding with skirt is defined as to be supplied to the recommendation information of user.
S204: recommendation information is supplied to user by server.
In above-mentioned steps S202, server, when identifying the content information of the literal types such as the commodity sign of doing shopping in the image of receipt, goods amount and time buying, can adopt aforesaid embodiment one ~ embodiment three to identify.Concrete, server can adopt embodiment one to identify the commodity sign of doing shopping in the image of receipt, adopts the goods amount in the image of embodiment two identification shopping receipt, adopts the time buying in the image of embodiment three identification shopping receipt.
Wherein, when adopting the embodiment two recognition value amount of money, predeterminable keyword is " unit price ", " receivable ", " amount of money " etc., and the word of specified type is the word of numeric type.After then server obtains the image of shopping receipt, identify the word in the image of shopping receipt, and determine that the participle " cargo price " be made up of the word identified comprises predetermined keyword " unit price ", and the word be positioned at after this participle " cargo price " is the word " 2.00 " of numeric type, thus determine word " 2.00 " goods amount for identifying of this numeric type.
Further, suppose to comprise a more than predetermined keyword in the participle that the word identified is formed, then according to the priority of each predetermined keyword, can determine that the word of the numeric type after the predetermined keyword that priority is the highest is the goods amount identified.
General, accurately can reflect that the priority of the keyword of goods amount is in descending order: unit price, receivable, the amount of money.This is because, for a shopping receipt, " unit price " reflects the actual value of each part commodity, its priority is the highest, and " receivable " reflects the total charge of all commodity on shopping receipt, " amount of money " generally refers to that user consigns to the face amount sum of the bank note of shop-assistant for this shopping, and this amount of money may be greater than the actual receivable amount of money.Therefore, in upper example, if the participle that the word that server identifies from the image of shopping receipt is formed comprises " cargo price " and " receivable " these two participles, then according to the priority of predetermined keyword " unit price " and " receivable ", the word of the numeric type after " cargo price " is only defined as the goods amount identified by server.
In the embodiment of the present application, after server to identify the content information of literal type in bill images by step S202, determine that the method to the recommendation information of user to be supplied can comprise several as follows by step S203:
Method A:
When the content information that server identifies from bill images at least comprises commodity sign and time buying, determine to be supplied to the recommendation information of user according to this commodity sign and time buying, specifically comprise: when not comprising the commodity sign of default easy consumption type in the commodity sign that server identifies, in historical record, determine the historical time that the commodity sign of the described easy consumption type of the last record is corresponding, if described historical time is greater than for the elapsed time preset corresponding to the commodity sign of described easy consumption type to the time span of the time buying identified, then merchandise news corresponding for the commodity sign of the easy consumption type of preserving is defined as to be supplied to the recommendation information of user, when comprising the commodity sign of default easy consumption type in the commodity sign identified, is recorded as the historical time that described user buys the commodity of described easy consumption type the time buying identified.
Adopt above-mentioned example, suppose that the commodity sign of the easy consumption type preset comprises toothpaste, toilet paper, liquid detergent.The elapsed time preset for toothpaste, toilet paper, the easy commodity sign consuming type of liquid detergent these three kinds is respectively t11, t12 and t13.Then the commodity sign of server in the image identifying this shopping receipt is skirt, swimming suit, toothpaste and slippers, and after identifying time buying t1, determine not comprise in the four kinds of commodity signs identified commodity sign toilet paper and the liquid detergent of easily consumption type, therefore, server is in historical record, the historical time finding this user last time purchase toilet paper is t2, and determine that the elapsed time preset for this toilet paper is t12, thus, judge whether historical time t2 is greater than elapsed time t12 to the time span of the time buying t1 identified, if be greater than, then illustrate that this user is necessary to buy toilet paper, the merchandise news of the toilet paper prestored is defined as to be supplied to the recommendation information of user.Similar, the easy consumption type of article liquid detergent that this is not bought for user, server also can carry out similar process to determine the recommendation information that will provide to user, just repeats no longer one by one here.
It should be noted that, the historical record described in the embodiment of the present application is: when buying commodity before user, is uploaded onto the server by corresponding bill images, the content information in the bill images that server identification is also recorded.
Further, when presetting elapsed time for the commodity easily consuming type, this frequency easily consuming the commodity of type can be bought according to this user of record to set, be specially: buy according to this user of record the historical time that this easily consumes the commodity of type in advance at every turn, determine the time interval of adjacent two historical times, the mean value in each time interval is set to the elapsed time that this easily consumes the commodity of type.
Adopt above-mentioned example, illustrate for the elapsed time of default toilet paper, the historical time of the toilet paper that this user of hypothetical record buys at every turn is t121, t122, t123, t124, then server determines the time interval t122-t121, t123-t122, t124-t123 of adjacent two historical times, the mean value (t122-t131+t123-t122+t124-t123)/3 in each time interval is set to the elapsed time t12 of toilet paper.
In addition, in said method A, when not comprising the commodity sign of default easy consumption type in the commodity sign that server identifies from the bill images that this gets, if it is determined that when historical time corresponding to the commodity sign of the described easy consumption type of the last record is less than elapsed time corresponding to the commodity sign of this easy consumption type to the time span of time buying (this time buying is the time buying that server identifies from the bill images that this gets), also can monitor the time span of this historical time to current time, until when historical time is greater than elapsed time corresponding to the commodity sign of this easy consumption type to the time span of current time, again the described merchandise news determined is supplied to user.
Method B:
When the content information that server identifies from bill images at least comprises commodity sign, determine to be supplied to the recommendation information of user according to this commodity sign, specifically comprise: according to the incidence relation between each commodity sign preset, in the commodity sign that respectively prestores, search the commodity sign that prestores be associated with the commodity sign identified, merchandise news corresponding for the commodity sign that prestores found is defined as to be supplied to the recommendation information of user.
Adopt above-described embodiment, the incidence relation between each commodity sign that setting is preset is: swimming suit, swimming cap and swimming trunks three are associated; Both toothpaste and toothbrush are associated.Then the commodity sign of server in the image identifying this shopping receipt is after skirt, swimming suit, toothpaste and slippers, according to the incidence relation of swimming suit, swimming cap and swimming trunks three, at the swimming cap and swimming trunks that to search in commodity sign and be associated with this swimming suit of prestoring, the swimming cap found and the corresponding respectively merchandise news of swimming trunks are defined as to be supplied to the recommendation information of user.
Method C:
When the content information that server identifies from bill images at least comprises commodity sign and time buying, determine to be supplied to the recommendation information of user according to this commodity sign and time buying, specifically comprise: according to each time period divided in advance, determine the time period at current time place, when the commodity sign determining to identify be time period at described current time place corresponding prestore commodity sign time, merchandise news corresponding for each commodity sign that the prestores corresponding time period at current time place is defined as to be supplied to the recommendation information of user.
Such as, divided for the first ~ four time period in advance, each time period is the season in 1 year, preset first time period (namely, the first quarter) the corresponding commodity sign that prestores comprises sweater, long woolen underwear, cotton knitwear, second time period (namely, the second quarter) the corresponding commodity sign that prestores comprises skirt, swimming suit, swimming trunks, cooling slippers, sun-screening umbrella, 3rd time period (namely, the third season) the corresponding commodity sign that prestores comprises autumn clothing, autumn trousers, woollen sweater, autumn footwear, 4th time period (namely, fourth quarter) the corresponding commodity sign that prestores comprises down jackets, eider down vest, cotton jacket, cotton-padded shoes.
Suppose that current time is on May 3rd, 2014, then the commodity sign of server in the image identifying this shopping receipt is skirt, swimming suit, toothpaste and slippers, and to identify current time be after 3 days Mays in 2014, the time period determining current time (on May 3rd, 2014) place was the second time period, and determine the commodity sign skirt that identifies, swimming suit is the commodity sign that prestores corresponding to this second time period, therefore, server determines that the merchandise news that each commodity sign that prestores corresponding to the second time period is corresponding is to be supplied to the recommendation information of user, also be, by skirt, swimming suit, swimming trunks, cooling slippers, the merchandise news that sun-screening umbrella is corresponding is defined as to be supplied to the recommendation information of user.
Certainly, in upper example, in view of user have purchased skirt and swimming suit, therefore, server also can by corresponding for this second time period, the merchandise news that each commodity sign except the commodity sign identified is corresponding is defined as to be supplied to the recommendation information of user, also, is defined as to be supplied to the recommendation information of user by swimming trunks, cooling slippers, merchandise news that sun-screening umbrella is corresponding.
Method D:
When the content information that server identifies from bill images at least comprises commodity sign and goods amount, determine to be supplied to the recommendation information of user according to this commodity sign and goods amount, specifically comprise: prestoring in commodity sign, determine the prestore commodity sign identical with the commodity sign identified, when goods amount corresponding to the commodity sign that prestores determined is lower than the goods amount that the commodity sign identified is corresponding, merchandise news corresponding for the commodity sign that prestores determined is defined as to be supplied to the recommendation information of user.
Concrete, commodity sign of supposing to prestore comprises skirt and swimming suit, and goods amount corresponding to the commodity sign that respectively prestores is respectively P11 and P12.Then the commodity sign of server in the image identifying this shopping receipt is skirt, swimming suit, toothpaste and slippers, and after the goods amount identifying skirt, swimming suit, toothpaste and slippers is respectively P1, P2, P3 and P4, relatively the size of P1 and P11, compares the size of P2 and P12.Suppose P1>P11, P2<P12, then illustrate that the merchandise news of the skirt preserved in server is more preferential, therefore merchandise news corresponding for the skirt of preservation is defined as to be supplied to the recommendation information of user.
Further, in said method A ~ method D, the commodity sign that prestores in server is predetermined, for diverse ways, is described as follows:
In said method A, for a certain commodity sign, whether server using this commodity sign as the method for the commodity sign of the easy consumption type prestored can be: the bill images uploaded according to the user obtained in the first preset time period, determine the purchase number of times of commodity in the first setting-up time section that this commodity sign is corresponding, relatively this purchase number of times and first sets threshold value, if buy number of times to be greater than the first setting threshold value, then using the commodity sign of this commodity sign as the easy consumption type prestored, otherwise, this commodity sign is defined as the commodity sign of non-easy consumption type.Concrete, server according to the commodity sign identified in the bill images uploaded from each user and time buying, can determine the purchase number of times of commodity in the first setting-up time section that this commodity sign is corresponding.
In said method B, server determines that the method that whether there is incidence relation between multiple commodity sign can be: the bill images uploaded according to the user obtained in the second preset time period, determine that the plurality of commodity sign comes from the probability of same bill images, if this probability is more than the second predetermined threshold value, then determine to there is incidence relation between the plurality of commodity sign, server then can be set up the incidence relation between the plurality of commodity sign and preserve, otherwise, determine to there is not incidence relation between the plurality of commodity sign.
Such as, the image of the shopping receipt that all users obtained according to (i.e. the second preset time period in) in month upload, determine that swimming suit, swimming cap and swimming trunks come from the probability of the image of same shopping receipt more than the second predetermined threshold value, then determine swimming suit, between swimming cap and swimming trunks three, there is incidence relation.
In said method C, server is for each time period divided in advance, the method setting the corresponding commodity sign that prestores for this time period divided in advance is specifically as follows: the bill images that server is uploaded according to the user obtained in the 3rd preset time period (during as a year), determine that the time buying is within this time period divided in advance, and buy number of times the prestore commodity sign corresponding more than the commodity of the 3rd setting threshold value, as the commodity sign that prestores that this time period divided in advance is corresponding.
It should be noted that, above-mentioned method A ~ method D can be used alone, and also can be combined several for wherein certain.Such as, suppose the time period at current time place be the 4th time period (as, the 4th season in 1 year), then the commodity sign of server in the image identifying this shopping receipt is swimming suit, and after determining that the time period at current time place was the 4th time period, first can search according to method B the commodity sign that prestores be associated with swimming suit in the commodity sign that respectively prestores, the set formed by the commodity sign that respectively prestores found is as the first set, the commodity sign that respectively prestores that the 4th time period at current time place is corresponding is determined again according to method C, the set formed by the commodity sign that respectively prestores determined is as the second set, finally merchandise news corresponding to the comprised commodity sign that prestores is concentrated to be supplied to user with the second intersection of sets the first set, if this common factor is empty, then not for user provides merchandise news.
In addition, in information providing method as above, are all the images obtaining user's collection with server, and are that example is described by the content information in this server recognition image, certainly, also can be carried out the content information in recognition image by terminal.Also be, after terminal gathers image, the content information (recognition methods can be any one or a few combination of above-mentioned embodiment one ~ tri-kind) in this image is identified by this terminal, and directly the content information identified can be sent to server, recommendation information to be supplied is determined according to the content information received again by server, finally the recommendation information determined is returned to this terminal, just repeat no longer one by one here.
Be the information providing method that the embodiment of the present application provides above, based on same thinking, the embodiment of the present application additionally provides a kind of information provider unit, as shown in Figure 3, comprising:
Image collection module 11, for obtaining the image that user gathers;
Content information identification module 12, for identifying the content information in described image;
Recommendation information determination module 13, to be supplied to the recommendation information of user for determining according to described content information;
Recommendation information provides module 14, for described recommendation information is supplied to described user.
Optionally, described image comprises bill images;
Described content information comprises: commodity sign, goods amount and at least one in the time buying.
Optionally, described content information identification module 12 specifically for:
Identify the word in described image, and determine the participle that is made up of the word identified;
When comprising predetermined keyword in the participle determined, described participle is defined as the content information identified; Or
When comprising predetermined keyword in the participle determined, and when the word be positioned at after the participle determined is the word of specified type, the word of described specified type is defined as the content information identified; Or
When the word that the word identified is specified type, and when the participle be made up of the word of described specified type falls into default set, the word of described specified type is defined as the content information identified.
Optionally, recommendation information determination module 13 specifically for:
When not comprising the commodity sign of default easy consumption type in the commodity sign identified, in historical record, determine the historical time that the commodity sign of the described easy consumption type of the last record is corresponding, if described historical time is greater than for the elapsed time preset corresponding to the commodity sign of described easy consumption type to the time span of the time buying identified, then merchandise news corresponding for the commodity sign of the easy consumption type of preserving is defined as to be supplied to the recommendation information of user;
When comprising the commodity sign of default easy consumption type in the commodity sign identified, is recorded as the historical time that described user buys the commodity of described easy consumption type the time buying identified.
Optionally, buy the historical time of the commodity of described easy consumption type in advance according to the described user of record at every turn, determine the time interval of adjacent two historical times;
The mean value in each time interval is set to the elapsed time of the commodity of described easy consumption type.
Optionally, recommendation information determination module 13 specifically for:
According to the incidence relation between each commodity sign preset, in the commodity sign that respectively prestores, search the commodity sign that prestores be associated with the commodity sign identified, merchandise news corresponding for the commodity sign that prestores found is defined as to be supplied to the recommendation information of user.
Optionally, recommendation information determination module 13 specifically for:
According to each time period divided in advance, determine the time period at current time place;
When the commodity sign determining to identify be time period at described current time place corresponding prestore commodity sign time, merchandise news corresponding for each commodity sign that the prestores corresponding time period at current time place is defined as to be supplied to the recommendation information of user.
Optionally, recommendation information determination module 13 specifically for:
Prestoring in commodity sign, determining the prestore commodity sign identical with the commodity sign identified;
When goods amount corresponding to the commodity sign that prestores determined is lower than the goods amount that the commodity sign identified is corresponding, merchandise news corresponding for the commodity sign that prestores determined is defined as to be supplied to the recommendation information of user.
The information providing method that the application provides, comprising: obtain the image that user gathers; Identify the content information in described image; Determine to be supplied to the recommendation information of user according to described content information; Described recommendation information is supplied to described user.In the embodiment of the present application, the recommendation information provided for user determines according to the content information in the image of user's collection, instead of to determine according to a large amount of on-line operation records of this user, therefore, even if user infrequently carries out on-line operation, also by the content information in the image of user's collection, determine the behavioural habits of user, and then be defined as the recommendation information that user provides, thus effectively can rise to the accuracy of the recommendation information that user provides.
Also it should be noted that, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, commodity or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, commodity or equipment.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, commodity or the equipment comprising described key element and also there is other identical element.
It will be understood by those skilled in the art that the embodiment of the application can be provided as method, system or computer program.Therefore, the application can adopt the form of complete hardware embodiment, completely software implementation or the embodiment in conjunction with software and hardware aspect.And the application can adopt in one or more form wherein including the upper computer program implemented of computer-usable storage medium (including but not limited to magnetic disk memory, CD-ROM, optical memory etc.) of computer usable program code.
The foregoing is only the embodiment of the application, be not limited to the application.To those skilled in the art, the application can have various modifications and variations.Any amendment done within all spirit in the application and principle, equivalent replacement, improvement etc., within the right that all should be included in the application.
Claims (16)
1. an information providing method, is characterized in that, comprising:
Obtain the image that user gathers;
Identify the content information in described image;
Determine to be supplied to the recommendation information of user according to described content information;
Described recommendation information is supplied to described user.
2. the method for claim 1, is characterized in that, described image comprises bill images; Described content information comprises: commodity sign, goods amount and at least one in the time buying.
3. the method for claim 1, is characterized in that, identifies the content information in described image, specifically comprises:
Identify the word in described image, and determine the participle that is made up of the word identified;
When comprising predetermined keyword in the participle determined, described participle is defined as the content information identified; Or
When comprising predetermined keyword in the participle determined, and when the word be positioned at after the participle determined is the word of specified type, the word of described specified type is defined as the content information identified; Or
When the word that the word identified is specified type, and when the participle be made up of the word of described specified type falls into default set, the word of described specified type is defined as the content information identified.
4. method as claimed in claim 2, is characterized in that, describedly determines to be supplied to the recommendation information of user according to content information, specifically comprises:
When not comprising the commodity sign of default easy consumption type in the commodity sign identified, in historical record, determine the historical time that the commodity sign of the described easy consumption type of the last record is corresponding, if described historical time is greater than for the elapsed time preset corresponding to the commodity sign of described easy consumption type to the time span of the time buying identified, then merchandise news corresponding for the commodity sign of the easy consumption type of preserving is defined as to be supplied to the recommendation information of user;
When comprising the commodity sign of default easy consumption type in the commodity sign identified, is recorded as the historical time that described user buys the commodity of described easy consumption type the time buying identified.
5. method as claimed in claim 4, it is characterized in that, elapsed time preset by the commodity for described easy consumption type, specifically comprises:
Buy the historical time of the commodity of described easy consumption type in advance according to the described user of record at every turn, determine the time interval of adjacent two historical times;
The mean value in each time interval is set to the elapsed time of the commodity of described easy consumption type.
6. method as claimed in claim 2, is characterized in that, describedly determines to be supplied to the recommendation information of user according to content information, specifically comprises:
According to the incidence relation between each commodity sign preset, in the commodity sign that respectively prestores, search the commodity sign that prestores be associated with the commodity sign identified, merchandise news corresponding for the commodity sign that prestores found is defined as to be supplied to the recommendation information of user.
7. method as claimed in claim 2, is characterized in that, it is characterized in that, describedly determines to be supplied to the recommendation information of user according to content information, specifically comprises:
According to each time period divided in advance, determine the time period at current time place;
When the commodity sign determining to identify be time period at described current time place corresponding prestore commodity sign time, merchandise news corresponding for each commodity sign that the prestores corresponding time period at current time place is defined as to be supplied to the recommendation information of user.
8. method as claimed in claim 2, is characterized in that, it is characterized in that, describedly determines to be supplied to the recommendation information of user according to content information, specifically comprises:
Prestoring in commodity sign, determining the prestore commodity sign identical with the commodity sign identified;
When goods amount corresponding to the commodity sign that prestores determined is lower than the goods amount that the commodity sign identified is corresponding, merchandise news corresponding for the commodity sign that prestores determined is defined as to be supplied to the recommendation information of user.
9. an information provider unit, is characterized in that, comprising:
Image collection module, for obtaining the image that user gathers;
Content information identification module, for identifying the content information in described image;
Recommendation information determination module, to be supplied to the recommendation information of user for determining according to described content information;
Recommendation information provides module, for described recommendation information is supplied to described user.
10. device as claimed in claim 9, it is characterized in that, described image comprises bill images;
Described content information comprises: commodity sign, goods amount and at least one in the time buying.
11. devices as claimed in claim 9, is characterized in that, described content information identification module specifically for:
Identify the word in described image, and determine the participle that is made up of the word identified;
When comprising predetermined keyword in the participle determined, described participle is defined as the content information identified; Or
When comprising predetermined keyword in the participle determined, and when the word be positioned at after the participle determined is the word of specified type, the word of described specified type is defined as the content information identified; Or
When the word that the word identified is specified type, and when the participle be made up of the word of described specified type falls into default set, the word of described specified type is defined as the content information identified.
12. devices as claimed in claim 10, is characterized in that, described recommendation information determination module specifically for:
When not comprising the commodity sign of default easy consumption type in the commodity sign identified, in historical record, determine the historical time that the commodity sign of the described easy consumption type of the last record is corresponding, if described historical time is greater than for the elapsed time preset corresponding to the commodity sign of described easy consumption type to the time span of the time buying identified, then merchandise news corresponding for the commodity sign of the easy consumption type of preserving is defined as to be supplied to the recommendation information of user;
When comprising the commodity sign of default easy consumption type in the commodity sign identified, is recorded as the historical time that described user buys the commodity of described easy consumption type the time buying identified.
13. devices as claimed in claim 12, is characterized in that, buy the historical time of the commodity of described easy consumption type in advance at every turn, determine the time interval of adjacent two historical times according to the described user of record;
The mean value in each time interval is set to the elapsed time of the commodity of described easy consumption type.
14. devices as claimed in claim 10, is characterized in that, described recommendation information determination module specifically for:
According to the incidence relation between each commodity sign preset, in the commodity sign that respectively prestores, search the commodity sign that prestores be associated with the commodity sign identified, merchandise news corresponding for the commodity sign that prestores found is defined as to be supplied to the recommendation information of user.
15. devices as claimed in claim 10, is characterized in that, described recommendation information determination module specifically for:
According to each time period divided in advance, determine the time period at current time place;
When the commodity sign determining to identify be time period at described current time place corresponding prestore commodity sign time, merchandise news corresponding for each commodity sign that the prestores corresponding time period at current time place is defined as to be supplied to the recommendation information of user.
16. devices as claimed in claim 10, is characterized in that, described recommendation information determination module specifically for:
Prestoring in commodity sign, determining the prestore commodity sign identical with the commodity sign identified;
When goods amount corresponding to the commodity sign that prestores determined is lower than the goods amount that the commodity sign identified is corresponding, merchandise news corresponding for the commodity sign that prestores determined is defined as to be supplied to the recommendation information of user.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410433045.8A CN105447708A (en) | 2014-08-28 | 2014-08-28 | Information offering method and information offering device |
| HK16109770.9A HK1221807A1 (en) | 2014-08-28 | 2016-08-16 | Method and device for providing information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410433045.8A CN105447708A (en) | 2014-08-28 | 2014-08-28 | Information offering method and information offering device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105447708A true CN105447708A (en) | 2016-03-30 |
Family
ID=55557844
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410433045.8A Pending CN105447708A (en) | 2014-08-28 | 2014-08-28 | Information offering method and information offering device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105447708A (en) |
| HK (1) | HK1221807A1 (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107967620A (en) * | 2017-10-30 | 2018-04-27 | 北京火星盒子网络信息技术有限公司 | Information processing method and server |
| CN108446937A (en) * | 2018-01-18 | 2018-08-24 | 大象慧云信息技术有限公司 | Recommend method and device in invoice purchase and sale side based on collaborative filtering |
| CN108734494A (en) * | 2017-04-20 | 2018-11-02 | 知而行（上海）营销咨询有限公司 | Electronic ticket distribution method based on image recognition |
| CN108900629A (en) * | 2018-07-25 | 2018-11-27 | 西安艾润物联网技术服务有限责任公司 | A kind of service push method and device |
| CN109416693A (en) * | 2016-06-24 | 2019-03-01 | 塞吉尼亚株式会社 | Server unit, terminal installation and information processing method |
| CN109426993A (en) * | 2017-08-24 | 2019-03-05 | 上海淘票儿信息科技有限公司 | Method and system for recommending viewing information |
| CN109637037A (en) * | 2018-12-20 | 2019-04-16 | 深圳推推应用技术有限公司 | A kind of management method and mobile terminal of shopping cart |
| CN110189195A (en) * | 2019-05-14 | 2019-08-30 | 北京三快在线科技有限公司 | A kind of method and device of commercial product recommending |
| CN111291211A (en) * | 2018-11-21 | 2020-06-16 | 上海博泰悦臻网络技术服务有限公司 | Target association recent dynamic recommendation and providing method, system, terminal and medium |
| CN112262402A (en) * | 2018-05-25 | 2021-01-22 | 三星电子株式会社 | Electronic device and method for controlling external electronic device |
| CN112989198A (en) * | 2021-03-30 | 2021-06-18 | 北京三快在线科技有限公司 | Push content determination method, device, equipment and computer-readable storage medium |
| WO2022052825A1 (en) * | 2020-09-11 | 2022-03-17 | 北京市商汤科技开发有限公司 | Data processing method and apparatus, device, and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102831200A (en) * | 2012-08-07 | 2012-12-19 | 北京百度网讯科技有限公司 | Commodity propelling method and device based on image character recognition |
| CN102855480A (en) * | 2012-08-07 | 2013-01-02 | 北京百度网讯科技有限公司 | Method and device for recognizing characters in image |
-
2014
- 2014-08-28 CN CN201410433045.8A patent/CN105447708A/en active Pending
-
2016
- 2016-08-16 HK HK16109770.9A patent/HK1221807A1/en unknown
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102831200A (en) * | 2012-08-07 | 2012-12-19 | 北京百度网讯科技有限公司 | Commodity propelling method and device based on image character recognition |
| CN102855480A (en) * | 2012-08-07 | 2013-01-02 | 北京百度网讯科技有限公司 | Method and device for recognizing characters in image |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109416693A (en) * | 2016-06-24 | 2019-03-01 | 塞吉尼亚株式会社 | Server unit, terminal installation and information processing method |
| CN108734494A (en) * | 2017-04-20 | 2018-11-02 | 知而行（上海）营销咨询有限公司 | Electronic ticket distribution method based on image recognition |
| CN109426993A (en) * | 2017-08-24 | 2019-03-05 | 上海淘票儿信息科技有限公司 | Method and system for recommending viewing information |
| CN107967620A (en) * | 2017-10-30 | 2018-04-27 | 北京火星盒子网络信息技术有限公司 | Information processing method and server |
| WO2019085401A1 (en) * | 2017-10-30 | 2019-05-09 | 北京火星盒子网络信息技术有限公司 | Information processing method and server |
| CN108446937A (en) * | 2018-01-18 | 2018-08-24 | 大象慧云信息技术有限公司 | Recommend method and device in invoice purchase and sale side based on collaborative filtering |
| CN112262402A (en) * | 2018-05-25 | 2021-01-22 | 三星电子株式会社 | Electronic device and method for controlling external electronic device |
| CN108900629A (en) * | 2018-07-25 | 2018-11-27 | 西安艾润物联网技术服务有限责任公司 | A kind of service push method and device |
| CN108900629B (en) * | 2018-07-25 | 2021-11-30 | 西安艾润物联网技术服务有限责任公司 | Service pushing method and device |
| CN111291211A (en) * | 2018-11-21 | 2020-06-16 | 上海博泰悦臻网络技术服务有限公司 | Target association recent dynamic recommendation and providing method, system, terminal and medium |
| CN109637037A (en) * | 2018-12-20 | 2019-04-16 | 深圳推推应用技术有限公司 | A kind of management method and mobile terminal of shopping cart |
| CN110189195A (en) * | 2019-05-14 | 2019-08-30 | 北京三快在线科技有限公司 | A kind of method and device of commercial product recommending |
| WO2022052825A1 (en) * | 2020-09-11 | 2022-03-17 | 北京市商汤科技开发有限公司 | Data processing method and apparatus, device, and storage medium |
| CN112989198A (en) * | 2021-03-30 | 2021-06-18 | 北京三快在线科技有限公司 | Push content determination method, device, equipment and computer-readable storage medium |
| CN112989198B (en) * | 2021-03-30 | 2022-06-07 | 北京三快在线科技有限公司 | Push content determination method, device, equipment and computer-readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| HK1221807A1 (en) | 2017-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105447708A (en) | Information offering method and information offering device | |
| CN103412937B (en) | A kind of search purchase method based on handheld terminal | |
| RU2544771C2 (en) | Inferring user-specific location semantics from user data | |
| CN102929939B (en) | The offer method and device of customized information | |
| CN107784059A (en) | For searching for and selecting the method and system and machine-readable medium of image | |
| CN105608117B (en) | Information recommendation method and device | |
| CN108319376B (en) | Input association recommendation method and device for optimizing commercial word promotion | |
| US20160328785A1 (en) | Device and method for targeted display of information | |
| WO2019056661A1 (en) | Search term pushing method and device, and terminal | |
| CN112150212A (en) | Shop recommendation method and device based on recommendation reason | |
| US20230214895A1 (en) | Methods and systems for product discovery in user generated content | |
| CN102682005A (en) | Method and device for determining preference categories | |
| WO2014015079A2 (en) | Method and apparatus of recommending clothing products | |
| CN106033415A (en) | A text content recommendation method and device | |
| TW201533588A (en) | Method and system for providing code scanning result information | |
| US20160210678A1 (en) | Systems for generating a global product taxonomy | |
| US20200226168A1 (en) | Methods and systems for optimizing display of user content | |
| CN108509497A (en) | Information recommendation method, device and electronic equipment | |
| CN103257962B (en) | information providing method and device | |
| CN107092609A (en) | A kind of information-pushing method and device | |
| US20230030560A1 (en) | Methods and systems for tagged image generation | |
| CN104978406A (en) | User behavior analysis method of Internet platform | |
| CN106202260A (en) | searching method, device and search engine | |
| CN106547365A (en) | The method and apparatus of commercial product recommending | |
| CN106294527A (en) | A kind of information recommendation method and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1221807 Country of ref document: HK |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160330 |
|
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: WD Ref document number: 1221807 Country of ref document: HK |