Systems and methods for tracking movement of individuals through a building receive, by one or more RF nodes disposed near an entrance to the building, RF signals from RF-transmitting mobile devices carried by persons near the entrance, capture an image of the persons while they are near the entrance, determine an identity and relative distance of each RF-transmitting mobile device from each RF node based on information associated with the RF signals received by that RF node, detect humans in the image, determine a relative depth of each human in the image, and assign the identity of each RF-transmitting mobile device to one of the humans detected in the image based on the relative distance of each RF-transmitting mobile device from each RF node and the relative depth of each human in the image, thereby identifying each individual who to be tracked optically as that individual moves throughout the building.

Embodiments herein provide for audience obfuscation during advertisement insertion in content. In one embodiment, a network element is communicatively coupled to a plurality of content distributers and to a plurality of advertisement decision servers (ADSs). The network element is operable to receive a first set of requests from a first of the content distributers for advertisements to be inserted into content provided by the content distributer, detect, in the requests, Personally Identifiable Information (PII) of subscribers to the first content distributer requesting the content, and remove the PII of the requests to anonymize the requests. The network element is operable to transfer the anonymized requests to a first of the ADSs, receive advertisements from the first ADS for placement into the content of the content distributers, direct the first content distributer to insert the advertisements into the requested content.

The subject technology predicts consumer sentiment based on demographics and other statics features of the consumer as well as dynamic features generated based on engagement of the consumer with previously presented targeted content. The sentiment predictions are used to recommend and generated new targeted content that is published to the consumer.

A method comprises: collecting data including combinations of images and accompanying text and user feedback of the combinations; building training data sets based on the collected data; training a plurality of neural networks using the training data; generating a creative feature vector based on a specified image using a first network of the trained plurality of neural networks; generating a target audience vector based on a specified target audience using a second network of the trained plurality of networks; generating a sequence of words based on the vectors using a third network of the plurality of trained neural networks; and transmitting the generated sequence of words and the specified image to the target audience over a network.

A system can associate interests and responsibilities that correspond to a user account with a tag, based on search data originated by the user account. The system can determine content to send to the user account based on the tag. The system can determine that an offering is first offered after sending the content to the user account. The system can determine that the user account has purchased the offering. The system can determine that a portion of a commission associated with the user account purchasing the offering is credited to sending the content to the user account based on the tag. The system can store an indication that the portion of the commission associated with the user account purchasing the offering is credited to sending the content to the user account based on the tag.

Novel system, methods, which include machine learning, and device for providing color and fashion recommendations, including for persons with visual impairment such as color blindness or complete blindness. Also, methods providing a data storage system for storing digital renditions of garments; providing a portable communication device to extract color and/or pattern from garments through use of a camera and at least one algorithm; providing a processor capable of accessing locally stored and/or remote information about or learning the preferred matching set of garments; assigning each garment in the set of garments a red-green-blue (RGB) value; providing a suitability ranking for matching compatibility of the garment or the set of garments; and providing recommendations for preferred matching garment or set of garments by organizing the garments in at least one queue selected from the group consisting of audial, tactile, visual or a combination thereof, wherein the individual imports garments or set of garments, through a series of photos or video, for bulk import into a virtual closet for the identification and assignment of type of garments or set of garments using human or computational methods.

An information processing system including a communication unit that acquires information related to an interaction between objects from a sensing device that detects the interaction between the objects, an emotion information database constructed by accumulating an evaluation value used when an emotion value of each object generated based on the information related to the interaction between the objects is calculated, a certification unit that certifies the sensing device and issues certification information to the sensing device, and an authentication unit that authenticates the information related to the interaction transmitted from the sensing device based on the certification information issued to the sensing device.

Aspects of the subject disclosure may include, for example, obtaining a first indication of a user of a communication device browsing a first website associated with a first vendor, obtaining a second indication of the user of the communication device browsing a second website associated with a second vendor that is different from the first vendor, determining, in accordance with the first indication and the second indication, that the first vendor and the second vendor operate in a common marketplace, responsive to the determining, obtaining an advertisement of the first vendor to provide to the communication device, and providing the advertisement to the communication device. Other embodiments are disclosed.

The present disclosure relates to improving recommendations for small shops on an ecommerce platform while maintaining accuracy for larger shops. The improvement is achieved by retraining a machine-learning recommendation model to reduce sample selection bias using a meta-learning process. The retraining process comprises identifying a sample subset of shops on the ecommerce platform, and then creating shop-specific versions of the recommendation model for each of the shops in the subset. A global parameter adjustment is calculated for the global model based on minimizing losses associated with the shop-specific models and increasing the probability of items being recommended from small shops. The latter is achieved by introducing regularizer terms for small shops during the meta-learning process. The regularizer terms serve to increase the probability that an item from a small shop will be recommended, thereby countering the sample selection bias faced by small-shop items.

A system that analyzes a user's communications to select a promotion that is presented to the user. The analysis may occur in two stages: a first stage analyzes a single communication from a user to determine whether the user is a potential target for a promotion; for potential targets, a second stage analyzes a history of communications from the user to generate a user profile. The system may then select a promotion based on the profile. The profile may include a set of profile tags that are considerably more detailed and granular than traditional demographic data; tags may for example indicate user affiliations with groups or ideas (such as religions or political parties), or user life cycle stages. Using these rich, detailed user profile tags, the system may achieve promotion response rates far above those from traditional advertising, which relies on cookies or simple demographic categories.