13 results about "Online advertising" patented technology

The application discloses a kind of multi-channel advertisement intelligent distribution control system and method based on deep learning, comprising the following steps: collecting multi-source advertisement delivery data, structured input dataset including user behavior sequence, advertisement content sequence and channel attribute sequence are constructed;Structured input dataset is encoded to feature vector, and three kinds of feature vectors are generated;Three kinds of feature vectors are input into ternary gate fusion structure, and fusion feature representation is generated;Fusion feature representation is input into improved PLE model, and click rate prediction value, conversion rate prediction value and budget cost prediction value are output;Strategy scoring function is constructed based on the above three prediction results, and the distribution score of candidate advertisement and channel combination is generated;According to distribution score ranking, the highest score combination is obtained, distribution instruction is generated and output to multi-channel advertisement system.The application improves the precision and cost control ability of advertisement distribution strategy under multi-target delivery, and is suitable for large-scale online advertisement delivery scene.

Online advertisers may demand compliance with certain standards for the content of emails and other digital content with which its advertisements may be associated. Emails may contain control objects directing users to a web page showing content related to oil change coupons. However, in some systems, methods, and processes, a small variable proportion of users may instead be directed to a different sequence of pages to conduct a compliance check on the contents of the email received by the user. The compliance check sequence of pages may offer the user an incentive to forward the email to a specified email address for subsequent compliance review. Delivery of an incentive may be conditioned on successful receipt of the forwarded email.

This application discloses a user profiling analysis method and server based on online advertising. The method includes: first, acquiring the target user's long-tail search keyword sequence and community interaction record sequence; merging the two sequences into a user behavior action unit chain and generating a set of behavior evolution relationship edges to form a user behavior semantic script; then, performing dependency parsing to generate an intent discourse structure containing a set of subordinate relationship identifiers and a set of causal modification relationship identifiers; performing hierarchical attention propagation operations to generate a hierarchical attention distribution map; and finally, extracting key plot nodes and semantic transition relationship nodes, mapping them to a set of advertising placement semantic anchor points, and generating a user profile feature cluster. By constructing a user behavior semantic script and parsing its intent discourse structure, this application can generate an accurate profile feature cluster that reflects the hierarchical organization and evolution of user interests.

Advertising methods, advertising systems, and advertising management methods. Technology is provided to enable online advertising operators to more easily manage outdoor advertising. The advertising method includes the following steps: a second management device obtains first advertising information representing an advertisement for a first store from a first management device, the first management device managing advertising information for multiple stores that actually exist, including the first store; a display device obtains the first advertising information from the second management device; the display device displays the advertisement for the first store in a public place at a size viewable by multiple people based on the first advertising information; the second management device obtains visitor access information for the first store; and the second management device sends information related to advertising costs to the first management device based on the access information, the visitor being a person who visited the first store based on the advertisement displayed in the public place.

Identifying potentially unfair practices in content and serving relevant advertisements is described. A total risk score associated with an online advertisement is determined by a computer system based on a first risk score and an issue risk score. The first risk score is based on online content associated with the online advertisement and the issue risk score is based on one or more risk attributes associated with the online advertisement. The total risk score is transmitted to a client computer, where the total risk score is displayed through a user interface in conjunction with the display of the online advertisement on the client computer.

The invention discloses a full-link generation type full-modal click rate prediction method and system, and relates to the field of online advertisement and recommendation systems.The method comprises the steps that multi-modal data and user features of texts, images and videos are preprocessed, and serialization, standardization and vectorization processing is carried out; aligning the multi-modal representation to a unified discrete potential space by using a multi-scale residual quantization encoder; generating personalized dynamic parameters of the target prediction network from the user state and scene context hierarchy based on the super network; generating a next item in which the user is interested and a click probability by adopting an autoregression learning normal form in combination with the personalized dynamic parameters; and performing end-to-end training by jointly optimizing the autoregression loss and the CTR prediction loss. According to the method, the nonlinear relationship and high-order feature interaction between the user and the article are effectively captured, and the accuracy of click rate prediction and the recommendation effect are remarkably improved by utilizing multi-modal information and a personalized parameter generation mechanism.

This technology provides online advertising businesses with the ability to more easily manage outdoor advertising. [Solution] The advertising method includes: a second management device acquiring first advertising information indicating an advertisement for the first store from a first management device that manages advertising information for multiple existing stores, including the first store; a display device acquiring the first advertising information from the second management device; the display device displaying the advertisement for the first store in a public place in a size that can be viewed by multiple people, based on the first advertising information; the second management device acquiring visit information of visitors to the first store; and the second management device transmitting information regarding advertising costs to the first management device based on the visit information, wherein the visitor is a person who visited the first store based on the advertisement for the first store displayed in the public place.

According to various embodiments, the online advertisement content platform providing system includes a server and an electronic device, wherein the server is configured to obtain at least one content of a specific product from the electronic device, wherein the at least one content includes an image of the specific product, acquire product attribute information of the specific product based on the at least one content, wherein the product attribute information corresponds to a plurality of product features, and provide, based on the product attribute information, recommendation information of a famous image identified based on a plurality of weight values related to the plurality of product features to the electronic device, wherein the recommendation information of the famous image includes at least one famous image, wherein the electronic device is configured to display a first screen including the at least one famous image and a predetermined number of famous features among the plurality of famous features based on the provided recommendation information of the famous image, wherein based on the specific product being the first product, the predetermined number of famous features includes a first group of famous features among the plurality of famous features, and based on the specific product being the second product, the predetermined number of famous features includes a second group of famous features among the plurality of famous features, and the second group of famous features may be at least partially different from the first group of famous features.

A performance internet marketing self-optimizing, anthropomorphic, artificial intelligence-based system and method to quote, compare and purchase products from a provider, including personal lines insurance, commercial lines insurance, property insurance and casualty insurance. The system and method collects information from an online advertising platform during the process and returns the collected information to the online advertising platform for improvement of the online advertising platform.

Example methods, apparatuses, and systems are presented that allows a user to make a secure purchase online, directly through accessing an online advertisement and without being redirected to multiple, cumbersome webpages to process different pieces of information to complete the transaction, while still leveraging existing e-commerce entities, such as existing payment platforms and existing ad / content networks. The present system includes a commerce ads engine (CA engine) that interfaces with the user through an app associated with the CA engine, a tokenization platform for authentication of the user, and a merchant providing relevant offer and check out information about a product being advertised in an online ad. The present system integrates new information embedded into online advertisements that interfaces with the CA engine and allows for the user to reliably provide authentication while purchasing the desired product, and without accessing or being redirected to multiple websites as is conventionally done.

The systems and techniques described herein relate to predicting user conversions in online advertising. Input data associated with user and advertisement features may be processed through neural networks to generate embedding representations or feature cross representations. A multi-task layer calculates probabilities associated with multiple user actions like clicks, page views, sign-ups, or purchases. Click-through and view-through conversion probabilities may be calculated to generate a score. The systems and techniques described herein perform predictions on multiple types of user actions despite data sparsity and negative transfer challenges, enhancing advertisement targeting and improving conversion metrics.

The invention discloses an online advertisement putting optimization method based on a knowledge graph, and the method comprises the following steps: obtaining online advertisement putting data, generating a corresponding graph structure, and building a putting optimization parameter set; retrieving candidate delivery configuration in the graph structure, and obtaining a delivery sub-graph; generating a candidate arm set and determining rule constraint information; executing feasible region reconstruction processing to obtain a feasible candidate arm set; calculating a structural uncertainty index for each candidate arm, and determining an arm-level implicit exploration parameter of the candidate arm; generating selection probability distribution of a feasible candidate arm set according to an EXP3-IX decision rule, and collecting delivery feedback data of a current delivery action; and retrieving an interpretation path set corresponding to the release feedback data, updating the accumulated loss value of the target candidate arm, and regenerating the selection probability distribution. According to the method, knowledge graph constraint reconstruction and gated EXP3-IX online updating are combined, and robust continuous putting optimization on the premise of compliance is achieved.

The present invention provides improved techniques for identifying online advertisement laundering and online advertisement injection that may occur during the serving of advertisements to browsers. For example, in one embodiment, a method of identifying fraudulent advertising activities may comprise obtaining information identifying a computer system of every party involved in supplying an advertisement to an end user device, determining whether a computer system of a first party and a computer system. of a second party provide websites and are not intermediary computer systems, determining whether the computer system of the first party provides a website of a type undesirable to advertisers, determining whether the computer system of the second party has usage and traffic characteristics indicative of not being an intermediary computer system, and determining a frequency of recurrence of the determined. conditions.