EP3289558A1 - Verfahren zum erzeugen von prioritätsdaten für produkte - Google Patents
Verfahren zum erzeugen von prioritätsdaten für produkteInfo
- Publication number
- EP3289558A1 EP3289558A1 EP16722563.0A EP16722563A EP3289558A1 EP 3289558 A1 EP3289558 A1 EP 3289558A1 EP 16722563 A EP16722563 A EP 16722563A EP 3289558 A1 EP3289558 A1 EP 3289558A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- products
- user
- product
- information
- priority data
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/535—Tracking the activity of the user
Definitions
- the present invention relates to a method for generating priority data for products to which information can be output.
- data for at least one product with which a defined user or a defined user device has interacted is stored at a first time.
- at least a first subset of the products are assigned priority data for the output of information to the defined user or to the defined user device for the products.
- a request for output of information about the products is captured by the defined user or device.
- the output of information on the products to the request then takes place depending on the stored priority data.
- the method is carried out in particular computer-aided or computer-implemented.
- the invention further relates to a method for operating a web page for e-commerce, in which this method is used for generating priority data for at least a first subset of the products, as well as a computer program product with program code for carrying out the method according to the invention.
- a large number of products are offered via a website.
- a user who calls up the website can not only search for these products on the website and obtain information about the products, but also buy these products via the website.
- data of the user are stored when the website is accessed. This user data is used when the user revisits the website in order to adapt the user's information display and search for specific products to the needs of the respective user. For example, the products last viewed and then not purchased by the user are stored. If the user then calls up the website again, they will be shown on the home page these last viewed products, in order to increase the likelihood that he buys these products when you visit the website again.
- the output of information about an ecommerce website is generally via priority data associated with the products. If the value of the priority data for a particular product is high, information about that product will appear more frequently, highlighted, or higher in a list than a product with a lower value of the associated priority data on the home page or in a search. The priority data should therefore be generated to best reflect the relevance of the associated product to a specific client computer or user.
- the present invention is therefore the technical problem of specifying a method for generating priority data for products of the type mentioned, which generates the priority data so that they the relevance of the associated products for a particular client computer or a defined user as accurately as possible depict.
- the inventive method is thus characterized in that at least in a part-time interval between the first and second time features of product interactions are detected, which were generated in an interaction by another user or another user device of the other user with the products of the first subset of products ,
- the stored priority data of the first subset of products is then changed depending on the particular detected characteristics of the product interactions.
- product in the sense of the invention refers in particular to objects which are offered to the user by a provider, in particular goods and services, such as physical objects or, for example, files such as music or video files, texts and images or web pages
- Product can also be understood in this document as a product category or a group of individual products with specific properties.
- an interaction of the defined user with at least one product is first detected. Based on this, priority data is assigned to a first subset of the products.
- This first subset may include, for example, only the product with which the user has interacted. It may also include some related products, such as in the same category. Furthermore, the subset may also be identical to the total amount of the products.
- the interaction of the user with a product has an influence on the priority data of other products or a defined quantity of products.
- the priority data of a product also depend on features of the product interactions generated by another user in interactions with the products in the time interval between a product interaction and a request for information to be output by the defined user.
- the user behavior of another user has an influence on the priority of a product for the defined user, ie both product interactions by the defined user and by other users are taken into account.
- the increasing or decreasing interest of other users in the products is taken into account over time.
- the priority data and the dependent output of information about the products will be adjusted accordingly. This makes it possible to better determine the relevance of a product for the defined user, so that the relevant user can be presented with more relevant products for the information output.
- the product interactions are in particular physical product interactions. These product interactions and / or their features are detected in particular by means of a detection unit, in particular a device for data acquisition. In a product interaction thus z. B. a data transfer, which is detected by the detection unit.
- the data stored in the method according to the invention for products with which an interaction has taken place contain information about which user or user device has interacted with this product.
- the priority data is stored for the later information output to the same user or user device.
- they are influenced by the user behavior of other users.
- a call of a website by another user or another user device of the other user is detected by means of a server.
- the features are captured by the product interactions in the part-time interval between the first and second times.
- data are stored at the first time for products to which information in a call a website by a client computer of the defined user or the defined user have been issued during the interaction.
- the first subset of the products are assigned the priority data for a re-invocation of the web page by this client computer or by this user.
- the request for informational output is a re-invocation of the webpage by this client computer or by that user.
- the output of information on the products when the website is called up again depends on the stored priority data.
- the first time is the call and the second time is the renewed call of the web page by the client computer or by the defined user.
- the priority data for the later call of the website by a defined client computer or by a user defined by the user behavior when calling the website by other users or another user device of the other user or another interaction by another user or other user device of the other user affected by the products.
- the product interactions of the defined user thus take place virtually on the website, that of the other user also via a website or in another way, as will be explained later.
- Calling or visiting the website in a broader sense means computer-implemented methods for requesting information and outputting it to a user.
- the method is thus not limited to the output of a web page in a browser, but also includes the output of data by means of a suitable application, such as on a mobile phone or mobile computer, or for example by a terminal in a shop and sending an e-mail ,
- the change of the priority data by the user behavior of the other users advantageously leads to the fact that the defined user can be given information on products that are as relevant for him as possible.
- information can be stored on which routine during the visit of the Website information about the product was issued. For example, it can be stored that the user has placed the product in an electronic shopping cart, but then did not buy the product. Similarly, it can be saved if the user has noted the product on an electronic watch list. Finally, it can be stored whether the user has obtained further information on the website on the website, for example by calling up a detailed product description.
- These data can also influence the value of the priority data for the assigned product, ie the priority of the product for the corresponding user or the assigned client computer.
- the product interactions and / or their features are detected in particular by means of a sensor. Since the priority data, which influence the output of information about the products to a corresponding request, depend on the detected product interactions or their characteristics, in the method according to the invention the output of the information is generated as a function of measuring signals of a sensor. In particular, a physical interaction of a user or a user device with a product is detected by means of the sensor and the measurement signals generated thereby are processed in such a way that features of the product interaction are generated.
- a spatial proximity between the products and the user and / or the user device is detected in the product interactions.
- physical product interactions can be detected, such as in a retail store or showrooms.
- the position of the user can be detected, for example by means of a camera or a tracking system, which locates the user on the basis of his user device.
- the user may be the defined user and / or the other user, and the user device may also be the user device of the defined user and / or the user device of the other user. In this way it can be determined, for example, whether the user is in the vicinity of the product, for example in front of a specific shelf.
- the features of the product interactions then depend in particular on the detected spatial proximity between the products and the user and / or the user device.
- a detection system can detect, for example, that the user is deliberately looking at a product. Furthermore, indirectly on the spatial proximity and a physical interaction, such as when the user pays for a product at a checkout facility. Furthermore, it can be detected when a user buys a product and later returns or exchanges it.
- a product interaction by the defined user or another user, or the corresponding user devices can be detected.
- product interactions may also be detected by unidentified users, such as the frequency of a particular type of interaction, such as how often the product is taken off the shelf or purchased.
- the user can identify himself, for example by login data of his user device, during a cashless payment process or through a customer card.
- the spatial position of the user relative to the spatial position of the products can be detected.
- the characteristics of the product interactions then depend on the detected relative position of the user to the products.
- a physical interaction of the user with the products can be detected.
- the characteristics of the product interactions then depend on the detected physical interaction of the user with one of the products.
- a code of one of the products can be detected by means of an electronic cash register system. The characteristics of the product interactions then depend on the detected code of one of the products.
- the product interactions transmit signals between the products and the user device by means of at least one transponder.
- radio systems with a limited range can detect whether the user is in close proximity to products.
- product interactions can be detected by the defined user or another user.
- signals can be transmitted, for example, by RFID (radio frequency identification) or NFC (near field communication).
- a product in particular on its packaging, comprise a transponder from which, for example, the user's mobile user device can read signals and identify the product.
- the position of the user can be detected by means of a transponder, for example in his customer card or at his shopping cart.
- both the product and the user can be identified by means of transponders, for example by a suitable reading device on one certain position in a showroom, at a shopping cart or at a cash register facility.
- the characteristics of the product interaction thus detected may include, for example, the identification of the product, the transmitted signal, and the time and place of the transmission.
- the captured product interactions may include both physical and virtual interactions.
- both the behavior of users of a web page and the behavior of users in a store may be used to determine or change the priority data of the products.
- the priority of products on the website may depend on how often the product is currently purchased in the store. As a result, the relevance of the products can advantageously be determined more accurately for the user.
- the time of generation of the feature of the product interaction is stored.
- the change of the stored priority data of the products then further depends on the time of generation of the respective feature of the product interaction.
- the time of the output of information on the product for another user or by another user device can also be stored and change the priority data.
- features of product interactions may be weighted differently depending on the time associated with the detection of the feature with respect to the influence of the features on the change of the priority data.
- the features of such product interactions may e.g. B. be more weighted, which are closer in time to the request for an output of information about the products by the defined user or the defined user device.
- features of such product interactions z. B. are weighted weaker, which are closer in time to the request for an output of information about the products by the defined user or the defined user device. For example, if a forerunner is interested in a product, this product will not be more interesting to that outrider than many other users are now buying.
- such features of product interactions that have been generated later are weighted more heavily than features of product interactions that were generated at an earlier point in time.
- features of product interactions which are subject to a temporal fluctuation the relevance of a product for the user when the website is called up can be better reflected in this way so that the user is advantageously provided with particularly relevant product information.
- features of product interactions that fluctuate over time include, for example, products with properties that are subject to seasonal fluctuations. For example, there are products that are particularly relevant for the winter time, other products are particularly relevant in the summer time. For example, by storing the time of generation of the feature of the product interaction, such as when another user invokes the web page, such seasonal variations may be taken into account.
- the features of the product interactions with the products include the frequency of interaction with these products by the other user or the other user device.
- the value of the priority data of the product increases when the product interaction, for example, the issuance of information about that product occurred frequently when the other client computer or the other user called up the web page.
- the features of the product interactions with the products include the frequency of a purchase of these products upon interactions by the other user or the other user device, such as when the web page is called by the other client computer or by the other user.
- the value of the priority data of the product increases particularly strong if the product has been purchased frequently by the other user or by his user device.
- the output of information about the products includes, in particular, the display of a representation or description of the product.
- this representation or description of the product may be part of a list of products.
- a list of product information when a list of product information is displayed, it has a ranking or order.
- the priority data of a product determines, in particular, the ranking of a product within the list.
- the ranking or the order of output of information on the products immediately after the request for output of information by the defined user or user device depends on the stored priority data for the products. For example, when calling up the website, in particular a so-called start page is displayed, which includes a list of product information.
- the ranking of this list is determined in this case, in particular, by the priority data of the products, so that priority data having a higher value is displayed in a higher rank in the list. For example, information about higher priority products may be displayed higher up in the list or highlighted.
- the method receives a search request for the search for products.
- a relevance evaluation is then performed on the relevance of the respective product for the search query.
- a hit list is then generated for the products, in which the ranking of the products depends on the relevance score and the priority data.
- the output of information about the products takes place according to the invention as a function of the stored priority values. Furthermore, the information output results from the corresponding request by the user or the user device. For example, this may be a specified interest of the user, such as a search request, for which a relevance rating of the products is first made independently of the individual user or the product interactions. To answer the specified interest, a relevance rating is performed, for example by means of the search terms. The relevance rating can take into account, for example, how many words of the search query are contained in the description text of a product or how quickly the respective product can be delivered. As methods for evaluating the relevance of search results, it is possible according to the invention to use methods such as those described in WO 2007/144199 A1 or WO 2012/025439 A1, the content of which is hereby incorporated by reference in the present specification.
- this relevance determination can be specified.
- personalization of the results can take place if, for example, the past product interactions of the user are taken into account.
- further information about the user for example the data of a user profile on a website, can also be included in the evaluation. For example, the age, gender or place of residence of the user can be included in the rating.
- the priority values are now adapted on the basis of the product interactions of other users.
- this makes it possible to make a time-dependent adaptation, for example, for an increasing and decreasing interest in a product in certain periods.
- the relevance determination on the basis of a search query is omitted and the priorities are based, for example, on past product interactions determined by the user and / or other users.
- the hit list generated in a conventional search query in this case thus depends not only on the relevance score to the search query, but also on the priority data, i. interim generated characteristics of product interactions by another user or user device.
- the hit list for the search request of the defined user can advantageously be improved.
- further parameters are included in generating the priority data.
- further data belonging to the products can be stored.
- the stored priority data of the products then depend on these further data.
- the other data may be, for example, an actual profit margin of a product and / or the current availability of the product.
- a measure of similarity to the products is also determined for the at least one product.
- the priority data of the first subset of the products is then further generated or changed depending on the similarity measure of the products with which a user or a user device has interacted.
- the user is the defined user and / or the other user
- the user device is also the user device of the defined user and / or the user device of the other user.
- the similarity of the products is determined in particular by means of strings which designate the products.
- the similarity of strings is examined.
- the similarity of two character strings is understood to mean a total value formed from the values of their non-overlapping fragments, preferably the sum of the adjusted values of their non-overlapping fragments, preferably divided by the total weight. If one understands the similarity of two strings in this way, then determining the similarity of two strings mathematically can be thought of as a weighted set packing problem, which involves finding subsets in a finite set that are not overlap.
- determining the similarity of two character strings may be considered to be the solution to a weighted set packing problem involving finding a selection of non-interleaved subsets of the highest total weight from a limited set.
- the weighted set packing problem is about finding from the set of fragments that can be formed with the two strings a selection of non-intersecting fragments that give the best possible score for the strings Similarity of the two strings overall result.
- the influence of an interaction with a product on the priority data of other products may depend on a weighting according to their similarity.
- “related" products that are more closely related may be more strongly influenced than less similar products.
- the similarity of the products may be determined, for example, by belonging to the same category or by common catchwords, and the products may be the same or similar.
- the similarity of the products to each other can also be determined by further methods known per se, such as a recommendation engine or a neural network, in which case the similarities need not be traceable to the unique characteristics of the products.
- the invention further relates to a method for operating a web page for e-commerce, in which an interaction with at least one product is detected by a defined user or a defined user device. Data is then stored for the product for which an interaction has been detected. Priority data is then generated for at least a subset of the products as described above. A call of the website by this user or his user device is then detected and information about the products is output depending on the stored priority data.
- the interaction with the at least one product is a call to the web page where information about the product is output, in particular an indication of a representation and / or description of the product.
- a client computer identification is transmitted when the website is called up, and it is determined from the transmitted client computer identification whether the website was called up by the client computer or another client computer.
- a user identification can be carried out and the user identification can be used to determine whether the website was called up by the user or another user.
- the interaction with the products is a spatial proximity between the products and the user and / or the user device as explained above.
- the invention further relates to a computer program product with program code for carrying out the method according to the method described above, when the program code is executed by a computer.
- Figure 1 shows the basic structure of a system for carrying out the
- FIG. 2 shows the structure of a server on which software is stored, with which the exemplary embodiment of the method according to the invention can be executed,
- Figure 3 shows the time evolution of the frequency of purchase of the product P-1 during one year
- Figure 4 shows the time evolution of the frequency of purchase of the product P-2 during one year.
- a server 1 On a server 1, software or program code for executing an e-commerce method is stored.
- the server 1 is connected to the Internet 2 or another suitable network.
- a user device 3-1 here a client computer 3-1, of a defined user N-1 is connected to the Internet 2.
- other user devices 3-2, 3-3, 3-4, 3-m in the illustrated case, client computers 3-2, 3-3, 3-4, 3-m, the user N-2 to Nm with the Internet 2 connected.
- the users N-1 to Nm can access the server 1 via the Internet 2 by means of the client computers 3-1 to 3-m and call a web page generated by the server 1 and in this way information on the client computers 3-1 to 3-m display.
- a website is understood here as the Internet presence of the e-commerce provider. The website can thus comprise several sub-web pages.
- the server 1 comprises a central control unit 5, which is coupled to a graphics unit 6.
- Graphics unit 6 generates advertisements for the website.
- the content of the display is predetermined by the central control unit 5.
- the graphics unit 6 accesses a database 7.
- the database 7 comprises data from a plurality of products P-1 to P-n.
- the product data include in particular representations and descriptions of the products P-1 to P-n.
- additional metadata about the products P-1 through P-n are stored, as will be explained later.
- the server 1 further comprises a search engine 8.
- the search engine 8 generates a hit list on search queries, in which an order of precedence lists information on products P-1 to Pn stored in the database 7.
- the control unit 5 is connected to a bidirectional interface 10 via which the server 1 is coupled to the Internet 2. Via the interface 2, the server 1 can transmit data for displaying the web page to the client computers 3-1 to 3-m. Further, inputs made by the client computers 3-1 to 3-m may be transmitted to the server 1 via the Internet 2.
- the server 1 comprises a detection unit 9.
- the detection unit 9 detects features of product interactions that are generated when the web page is called by another client computer 3-2 to 3-m or by another user N-2 to N-m. By means of the detection unit 9, an identification of another client computer 3-2 to 3-m and / or of another user N-2 to N-m is thus carried out. These features of product interactions are also stored in the database 7 as metadata for the products P-1 to P-n. Which features of product interactions are detected by the detection unit 9 will be explained later with reference to the embodiment of the method according to the invention.
- the database 7 contains priority data for the products P-1 to P-n as further metadata, which contain priority values for different users. These priority data can be changed by the various devices of the server 1, in particular by the central control unit 5, the search engine 8 and the detection unit 9. Details of how these priority data are generated and changed will be explained later with reference to the embodiment of the method according to the invention.
- server 1 An exemplary embodiment for operating a web page for e-commerce by means of server 1, which includes an embodiment for generating priority data for products P-1 to P-n, is described below.
- information about a plurality of products P-1 to Pn is stored in the database 7.
- a representation and a description of the product is included. It also stores attributes about the product and metadata that includes priority data for specific users or client computers.
- the user N-1 calls the web page of the server 1.
- a client computer identification is transmitted to the server 1, which stores it in the database 7.
- the time of the call of the web page by this client computer 3-1 is stored.
- an identification of the user N-1 is performed when the website is called up. For example, the user N-1 logs in by means of a password. Also, the time of logging in this user N-1 is stored in the database 7.
- the user N-1 via the client computer 3-1 in known manner information from the server 1 from and interacts with the products P-1 to P-n, to which he retrieves information about. For example, it can perform a search for different products P-1 to P-n of the e-commerce provider.
- the graphics unit 6 and the central control unit 5 the user N-1, for example, a hit list is displayed, which includes information on the different products P-1 to P-n of the e-commerce provider in a certain order.
- metadata about products P-1 to P-n are stored, which in particular comprise the information with which products P-1 to P-n the user N-1 has interacted.
- Such features of the product interactions include, for example, to which products P-1 to Pn information was output, which products P-1 to Pn the user N-1 has placed in an electronic shopping cart or added to an electronic shopping list, and which products P-1 until Pn the user bought N-1.
- priority data on the products P-1 through Pn and on the products P-1 through Pn are generated depending on the data acquired during the visit of the website User N-1 is generated and stored in the database 7.
- priority data is generated for a subset of the total quantity of the products P-1 to Pn. This subset may include exactly those products P-1 through Pn with which the user has interacted with N-1. However, in another example, it may also contain other products P-1 through Pn, which belong to about the same category or have similar properties as the products P-1 through Pn, with which the user has interacted with N-1.
- this subset can be determined by determining the similarity of the individual products P-1 to Pn in relation to the products P-1 to Pn with which the user N-1 has interacted by a similarity measure. This can also be done by using a neural network that determines similarity between products P-1 to Pn without necessarily having similar properties. As a method for determining the similarity measure further methods can be used, as described in WO 2007/144199 A1 or WO 2012/025439 A1.
- products P-1 to Pn are assigned a first priority value, to products P-1 to Pn, which the user N-1 has added to the electronic watch list, a second, assigned higher priority value and assigned to products P-1 to Pn, which has put the user N-1 in the electronic shopping cart, but has not bought, a third, even higher priority value assigned.
- the subset includes only the products P-1 to P-n with which the user has interacted with N-1.
- the other products P-1 to Pn are similar to the first product P-1 to Pn, for example, the same model of a product P-1 to Pn in a different color, in a different material or a similar product P-1 to Pn one other manufacturer.
- This priority data is used in a renewed call of the web page by the client computer 3-1, to which the client computer identification has been stored, or by the user N-1, which was determined on the basis of the user identification, in order to obtain information on these products. 1 to Pn, for example, on the homepage of the web page.
- the server 1 If, for example, the user N-1 calls the website again by means of the client computer 3-1, this is detected by the server 1 by means of the client computer identification or by means of the user identification.
- the central control unit 5 retrieves from the database 7 the priority data stored for this user N-1 or this client computer 3-1 for specific products P-1 to Pn and generates, by means of the graphics unit 6, a start page which displays a list with information about Products P-1 to Pn contains.
- the ranking of the products P-1 to Pn in this list depends on the priority data. For example, information about the products P-1 to Pn that the user N-1 put in the electronic shopping cart at the last time the website was called up, but did not purchase it, can be displayed with the highest rank.
- the next lower rank displays information on the products P-1 through Pn which the user N-1 has added to a watch list.
- information is displayed on the products P-1 to Pn which were only displayed to the user N-1 when the web page was last accessed.
- the rank of different products P-1 through Pn may be used, for example, to generate a promotional e-mail to the user N-1 in order to output information on the products P-1 to Pn in this way.
- the user is given N-1 information without specifying the desired information, for example by a search query. For example, the call of the web page on whose home page the list of information about products P-1 to Pn is located corresponds to such a non-specific information query.
- the relevance score added to the priority data for the products P-1 to Pn is carried out independently of the user N-1 , For example, the user searches for the term "cloak.” First, those products are searched for that correspond to the search term, and ordered according to their relevance, and the priority order for the products is used to change the ranking of the products, for example if the user N-1 Furthermore, interactions other users N-2 to Nm are considered and, for example, such a coat of high rank is displayed, which is currently being purchased very frequently.
- virtual product interactions take place on a web page, for example, by invoking information about a product P-1 through P-n, adding it to a watch list, or purchasing.
- the distribution of information about the products may be by other means, such as an ad, brochure or e-mail.
- product interactions it can then be evaluated, for example, that the user N-1 receives an e-mail with information about a product P-1 to Pn, calls it and / or via a link in the e-mail further information on a particular product P -1 calls up to Pn.
- information on a plurality of products P-1 through P-n is also stored here as the starting point for the method in the database 7.
- metadata are stored, which include priority data for certain users N-1 to N-m and client computers 3-1 to 3-m, respectively.
- the products P-1 to Pn are on the shelf in the showrooms and salesrooms of a retail store.
- P-1 to Pn Transponder attached such as RFID or NFC transponder, of which the identification of the individual products P-1 to Pn can be read by a suitable reader.
- the user N-1 is moving through the store, carrying with him a user device 3-1, such as a mobile phone 3-1, which includes means for reading out the transponders. If the user N-1 stands in front of the shelf in which the product P-1 to Pn is located, an application installed and executed on the mobile phone 3-1 may be in close proximity to the product P-1 to detect Pn (or to the RFID transponder) and transmit the features of this product interaction to the server 1.
- the use of the mobile phone 3-1 allows identification of the user N-1, for example by the transmission of login data to the server. 1 For example, it can be detected when and for how long the user N-1 dwells on the product P-1 to Pn. Furthermore, further product interactions can be detected, for example, whether the user N-1 takes the product P-1 to Pn from the shelf, whether he can display about the mobile phone 3-1 more information on the product P-1 to Pn or whether he places the product P-1 to Pn in his shopping cart. Further, it can be detected when the user N-1 buys the product P-1 to Pn and pays at the cash register facility of the store.
- the transponder When transponders are used to detect the product interaction, the transponder may be attached to the product P-1 to Pn, in particular to its packaging, but the user N-1 may also carry a transponder, for example in the form of a suitable user device. 1 or a transponder in a customer or credit card or on a shopping cart. Such a transponder associated with the user N-1 may be read out, for example, from facilities in the business premises to the position of the user, for example in front of the shelf with the product P-1 to Pn, or its interactions with products P-1 to Pn to capture.
- the detection of the product interactions can also be done by other means, for example by a camera system or a data goggles of the user N-1, for example, detect where the user N-1 is located, whether the user N-1 a product P-1 to Pn takes it off the shelf, places it in the shopping cart or buys it.
- a system may also identify or capture the user N-1 which views products P-1 through P-n of the user N-1.
- the camera system can be installed in the shop or mobile, for example, as a camera of a data glasses of the user N-1.
- the cash register device can detect purchase transactions as product interactions and transmit them to the server 1.
- priority data are generated on the products P-1 to Pn and for the user N-1 and stored in the database 7.
- products will P-1 to Pn, which the user N-1 took off the shelf, assigned a first priority value, to products to which the user has called N-1 additional information, assigned a second, higher priority value and to products P-1 until Pn, which the user N-1 has put into the shopping cart, but has not finally bought, assigned a third, even higher priority value.
- product interactions of other users N-2 to Nm can also be evaluated; for example, a large number of purchase transactions can be assigned a specific priority value.
- the priority data for other products P-1 through P-n may also be changed than just those with which user N-1 has interacted.
- the priority data of products P-1 through P-n may be changed based on the characteristics of interactions with similar products. For example, frequent purchase of winter coats may also increase the priority data of other products P-1 to P-n from the category "winter clothing".
- this priority data is used to display information about these products P-1 through P-n, for example, on the homepage of the website.
- the output of the information about the products P-1 to Pn is analogous to the method described above, wherein the ranking of the products P-1 to Pn in the displayed list also depends on the priority data obtained from the data on the physical product interactions were determined. For example, information about the products P-1 to P-n that the user N-1 put into the shopping cart when the website was last accessed, but did not purchase it, can be displayed with the highest rank.
- the user N-1 can also be shown information about the products P-1 to P-n in the physical space, such as the store. For example, a display device in the shop detects that the user N-1 is in front of it. Depending on the priority data of the products P-1 to P-n, the user is then displayed information that is relevant to him. For example, a consultant can also identify the user N-1 and retrieve information about it, information on which products P-1 to P-n are particularly relevant for the user N-1. In this case, the user N-1 is provided with the information through the conversation with the advisor.
- the user N-1 interacts during a conversation with the products P-1 to Pn:
- a product P-1 to Pn can be presented to the user N-1, for example as part of a consultation or a telephone call.
- the verbal transmission of information to the user corresponds to the display of information on the product P-1 to Pn on a website. It can, for example the interest of the user N-1 in the presented product P-1 to Pn are recorded. Further, it may be detected whether the user N-1 is requesting, for example, information about a particular product P-1 to Pn or buying a product P-1 to Pn.
- both physical product interactions, such as in a store, and virtual product interactions, such as on a web page, are captured and contribute to the determination of the priority data.
- the product interactions described above can also be detected analogously for other users N-2 to N-m and transmitted to the server 1.
- the product information can also be detected without an identification of the individual user.
- the frequency of a purchase may also be determined if the identity of individual buyers is unknown.
- this is the time interval between a call to the ecommerce web page described above and a reopening of the web page by the client computer 3-1 or by the user N-1:
- an interaction intensity for example the frequency of a specific interaction, as a function of time, is determined from the features of the product interactions.
- a smoothing method is used to distribute the features of the product interactions acquired at discrete points of time over surrounding periods. In particular, averaging can take place over a defined period of time, for example over a week, a month or a year.
- the interaction intensity at the time of the user N-1's earlier interaction with the product is compared with the interaction intensity at the time the website is called again. For this comparison, for example, the quotient of the two values can be formed in order to determine the current priority data for the user N-1.
- the frequency with which information about a product P-1 to P-n was displayed upon a call of the web page by another client computer 3-2 to 3-m or another user N-2 to N-m is determined. Furthermore, the frequencies are determined for a product from another user N-2 to N-m or by other client computers 3-2 to 3-m was added to the shopping cart or placed on a watch list or was purchased. These frequencies also belong to the features of the product interactions that are detected by the detection unit 9 in the time interval between the call and the renewed call of the web page by the client computer 3-1 or the user N-1. The greater this frequency, the more the value of the priority data is increased. Furthermore, the value of the priority data is even more increased when a product has been purchased from another user N-2 to Nn compared to the case where the product P-1 to Pn is placed only in the electronic shopping cart or on the electronic Watchlist has been set or has only been displayed.
- the time interval between the call and the re-invocation of the web page by the client computer 3-1 or by the user N-1 is at most considered. However, if this time interval is longer than a pre-defined one Is part-time interval, only acquisitions of features of product interactions in that part-time interval are taken into account prior to the re-invocation of the web page by the client computer 3-1 or by the user N-1.
- the part-time interval may include, for example, a period of one week, one month or four months. In this way seasonal or seasonal variations in the relevance of a product P-1 to Pn can be taken into account.
- the priority data of the products P-1 to P-n are not changed until the web page is called again by the client computer 3-1 or the user N-1. Rather, the priority data may also be continually changed if the features of the product interactions were detected due to the website's call by other users N-2 through N-m.
- the above-described physical product interactions by other users N-2 to N-m or other user equipment 3-2 to 3-m may contribute to changing the priority data.
- the priority data may also depend on the frequency in a store. For example, this may give a product P-1 to P-n, which is currently being purchased especially frequently by other users N-2 to N-m, a higher priority.
- the combination of data from physical and virtual product interactions may improve the accuracy of relevancy determination.
- current profit margins or availabilities of the product P-1 to P-n can also be stored in the database 7 for the products P-1 to P-n.
- the priority data of a product P-1 to P-n can then be increased additionally with a high profit margin or lowered with a low profit margin. Further, the priority data may be increased upon availability of the product P-1 to P-n, and even more at high stock of the product P-1 to P-n, or lowered in case of unavailability of the product P-1 to P-n.
- the changed priority data is then used to be stored on the client computer 3-1 e.g. to display a home page containing a list of information about products P-1 to P-n.
- the ranking of this information on the products P-1 to P-n in this case depends on the value of the priority data of the products P-1 to P-n. Products with high values of priority data are displayed higher and / or more emphasized than products P-1 through P-n with lower values of priority data.
- the priority data may also be taken into account by the search engine 8.
- the search engine 8 Upon a search, it performs a relevance score on the relevancy of a product P-1 through Pn for the search query.
- the hit list for the search query becomes then generated as a function of this relevance score and additionally in dependence on the priority data.
- Figures 3 and 4 an example of the evolution of the absolute frequency of purchase of a product P-1 (Figure 3) and a product P-2 ( Figure 4) during one year is shown.
- the numbers on the abscissa axis indicate the months of a year.
- the product P-1 is a winter article such as a down jacket
- the product P-2 is a summer article such as a swimsuit. It turns out that the product P-1 has been bought especially frequently in the winter months by other users N-2 to N-m. The product P-2, however, has been bought especially frequently in the summer months.
- the user N-1 calls the web page at time t1 in February and performs a search for various products. He puts the products P-1 and P-2 in his shopping cart, but does not buy these products P-1 and P-2 and leaves the website again. It is stored in the database 7 that this user N-1 has put the products P-1 and P-2 into the shopping cart at the time t1. Accordingly, the value of the priority data of these products P-1 and P-2 is set to a certain value for this user N-1.
- the priority data for other products are also changed by the interactions shown in FIGS. 3 and 4 for the products P-1 and P-2.
- the system may find a greater similarity to winter coats than to light summer jackets, for example, based on the product category.
- the relevance of P-1 increases, the relevance of winter coats may increase, while the less similar lightweight summer jackets remain unaffected.
- the priority data may also be used for other personalization applications.
- Such applications include, for example, product recommendation methods P-1 through Pn or methods of selecting products P-1 through Pn, which are offered in an individualized message, for example, an e-mail advertising letter.
- data is not stored for products P-1 to Pn to which information has been output upon a call of the web page by the client computer 3-1 or from the user N-1, but to information at a client computer 3-1 or have been issued to a user N-1.
- the output can also be made in other ways, for example via an e-mail, an advertising mail or the like.
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computer Hardware Design (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP15165337.5A EP3089097A1 (de) | 2015-04-28 | 2015-04-28 | Verfahren zum erzeugen von prioritätsdaten für produkte |
PCT/EP2016/059498 WO2016174142A1 (de) | 2015-04-28 | 2016-04-28 | Verfahren zum erzeugen von prioritätsdaten für produkte |
Publications (1)
Publication Number | Publication Date |
---|---|
EP3289558A1 true EP3289558A1 (de) | 2018-03-07 |
Family
ID=53274354
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP15165337.5A Ceased EP3089097A1 (de) | 2015-04-28 | 2015-04-28 | Verfahren zum erzeugen von prioritätsdaten für produkte |
EP16722563.0A Withdrawn EP3289558A1 (de) | 2015-04-28 | 2016-04-28 | Verfahren zum erzeugen von prioritätsdaten für produkte |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP15165337.5A Ceased EP3089097A1 (de) | 2015-04-28 | 2015-04-28 | Verfahren zum erzeugen von prioritätsdaten für produkte |
Country Status (4)
Country | Link |
---|---|
US (1) | US11170428B2 (de) |
EP (2) | EP3089097A1 (de) |
CN (1) | CN107743634A (de) |
WO (1) | WO2016174142A1 (de) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102017207992A1 (de) * | 2017-05-11 | 2018-11-15 | Homag Gmbh | Verfahren zur Überwachung eines Fertigungsprozesses |
EP3766034A1 (de) | 2018-03-16 | 2021-01-20 | Omikron Data Quality GmbH | Computerimplementiertes verfahren zum erzeugen einer vorschlagsliste und anlage zum erzeugen einer bestellliste |
CN110363613A (zh) * | 2019-05-28 | 2019-10-22 | 成都美美臣科技有限公司 | 一个电子商务网站获取潜在客户数据的方法 |
Family Cites Families (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2007144199A1 (de) | 2006-06-16 | 2007-12-21 | Omikron Data Quality Gmbh | Verfahren zum automatischen bewerten der ähnlichkeit von zwei zeichenketten, die in einem computer gespeichert sind |
US9606621B2 (en) * | 2006-07-28 | 2017-03-28 | Philips Lighting Holding B.V. | Gaze interaction for information display of gazed items |
CN101436186B (zh) * | 2007-11-12 | 2012-09-05 | 北京搜狗科技发展有限公司 | 一种提供相关搜索的方法及系统 |
US8589503B2 (en) * | 2008-04-04 | 2013-11-19 | Mcafee, Inc. | Prioritizing network traffic |
US8307308B2 (en) * | 2009-08-27 | 2012-11-06 | International Business Machines Corporation | Updating assets rendered in a virtual world environment based on detected user interactions in another world |
US20110270850A1 (en) | 2010-04-30 | 2011-11-03 | Microsoft Corporation | Prioritization of Resources based on User Activities |
EP2423830A1 (de) | 2010-08-25 | 2012-02-29 | Omikron Data Quality GmbH | Verfahren zum Suchen in einer Vielzahl von Datensätzen und Suchmaschine |
EP2730137B1 (de) * | 2011-07-05 | 2022-11-09 | HMD Global Oy | Verfahren und vorrichtung zur ressourcenaggregation in der drahtlosen kommunikation |
US20130275913A1 (en) * | 2012-04-13 | 2013-10-17 | Telefonaktiebolaget L M Ericsson (Pub) | Presentation of Items on a Display |
CN103390000B (zh) * | 2012-05-09 | 2016-07-06 | 中国移动通信集团公司 | 一种网页搜索方法及网页搜索系统 |
US10091323B2 (en) * | 2012-05-17 | 2018-10-02 | The Meet Group, Inc. | Social discovery feed for facilitating social exploration in social networking environments |
US20140058833A1 (en) * | 2012-08-22 | 2014-02-27 | Myworld, Inc. | Commerce System and Method of Controlling the Commerce System Using Bidding and Value Based Messaging |
US9965799B2 (en) * | 2012-12-12 | 2018-05-08 | Perch Interactive, Inc. | Apparatus and method for interactive product displays |
US10325285B1 (en) * | 2013-06-28 | 2019-06-18 | Groupon, Inc. | Predictive recommendation system |
US10496660B1 (en) * | 2013-09-09 | 2019-12-03 | Google Llc | Serving content items in content item slots based on a referral query location |
US10318986B1 (en) * | 2015-12-10 | 2019-06-11 | Thirstie Inc. | Method and system for curated product marketing |
-
2015
- 2015-04-28 EP EP15165337.5A patent/EP3089097A1/de not_active Ceased
-
2016
- 2016-04-28 EP EP16722563.0A patent/EP3289558A1/de not_active Withdrawn
- 2016-04-28 WO PCT/EP2016/059498 patent/WO2016174142A1/de active Application Filing
- 2016-04-28 CN CN201680024912.XA patent/CN107743634A/zh active Pending
- 2016-04-28 US US15/569,816 patent/US11170428B2/en active Active
Also Published As
Publication number | Publication date |
---|---|
EP3089097A1 (de) | 2016-11-02 |
WO2016174142A1 (de) | 2016-11-03 |
CN107743634A (zh) | 2018-02-27 |
US11170428B2 (en) | 2021-11-09 |
US20180130116A1 (en) | 2018-05-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
DE10196670B3 (de) | System und Verfahren zur Migration von Daten in einem elektronischen Handelssystem | |
DE112012003376T5 (de) | Seitenberichterstattung | |
DE10235804A1 (de) | System und Verfahren für das Ermöglichen eines Mehrelementbietens für das Beeinflussen einer Position in einer Suchergebnisliste die durch eine Suchmaschine eines Computernetzes erzeugt wird | |
DE10235812A1 (de) | Überwachung des Kontos einer Suchmaschine | |
DE10235429A1 (de) | System und Verfahren für das Bereitstellen eines Orts- und Preisschutzes in einer Suchergebnisliste die durch eine Suchmaschine eines Computernetzes erzeugt wird | |
DE10244974A1 (de) | Automatische Inserentenbenachrichtigung für ein System zum Bereitstellen eines Platz- und Preisschutzes in einer Suchergebnisliste, die durch eine Computernetzwerksuchmaschine erzeugt wird | |
DE202009019060U1 (de) | System zum Platzieren zielgerichteter Inhalte | |
US8166015B2 (en) | Web advertising management method | |
CN103886487A (zh) | 基于分布式的b2b平台的个性化推荐方法与系统 | |
WO2019175443A1 (de) | Computerimplementiertes verfahren zum erzeugen einer vorschlagsliste und anlage zum erzeugen einer bestellliste | |
CN107292648A (zh) | 一种用户行为分析方法及装置 | |
EP2827604A1 (de) | Verfahren, Server und Computerprogramm zum Optimieren von Onlinewerbung | |
US20030023511A1 (en) | Automated system and method for tracking online activity to initiating information source | |
WO2016174142A1 (de) | Verfahren zum erzeugen von prioritätsdaten für produkte | |
CA2713036A1 (en) | Object customization and management system | |
DE10245992A1 (de) | Vereinfachung der Bestellungseingabe | |
EP3340072B1 (de) | Verfahren zur durchführung einer selbsttätigen datenabfrage und programmprodukt | |
EP1484696B1 (de) | Verfahren zum Optimieren eines auf eine Netzwerkseite verweisenden Verweises | |
EP3096241A1 (de) | Verfahren zum durchsuchen einer datenbasis mit datensätzen | |
DE112017007652T5 (de) | Empfehlungseinrichtung | |
DE112021000337T5 (de) | Verfahren zum Extrahieren von Zielbenutzermerkmale, System zum Extrahieren von Zielbenutzermerkmale und Zielbenutzermerkmal-Extraktionsserver | |
DE202010017812U1 (de) | System zum Identifizieren von in Inhalt einzubeziehenden Begriffen | |
DE102014006951A1 (de) | Verfahren und Vorrichtung zum Aufbau von Kommunikation | |
DE19959142A1 (de) | Verfahren und Vorrichtung zum Übermitteln von Inhalten, insbesondere von Werbung | |
EP3076343A1 (de) | Verfahren zur zuordnung sprachlicher eingaben |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
17P | Request for examination filed |
Effective date: 20171020 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
AX | Request for extension of the european patent |
Extension state: BA ME |
|
DAV | Request for validation of the european patent (deleted) | ||
DAX | Request for extension of the european patent (deleted) | ||
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
17Q | First examination report despatched |
Effective date: 20200519 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN |
|
18D | Application deemed to be withdrawn |
Effective date: 20220611 |