Yahoo! Inc.

Go Daddy Operating Company, LLC

Anirban Kundu

Mark Risher

Vishwanath Tumkur Ramarao

Lei Zheng

Garrett Matsudaira

Stacy SteinKuller

Naveen Jamal

Miles Augustus Libbey, Iv

Mark Delany

Xiaopeng Xi

Bhasker Hariharan

Masumi Taketomi

Ana Alicia Grace

Joshua Ray Rossman

Angeley Mullins

Stephen Russell Baird

Hao Zheng

David Hikaru Nakayama

Jing Zhu

Sharat Narayan

Stanley Ke Wei

Bruce Hui-der Chu

Miles A Libbey

In some embodiments, the present disclosure describes a method that includes at least: receiving, by a processor, audience data identifying candidate audience for a website content; generating a plurality of audience groups based on the audience data; receiving user identifying data of a user; generating personalized online content for the user; tagging the personalized online content with a plurality of dynamic tokens to form trackable personalized online content; transmitting instructions to display the trackable personalized online content on a screen of a computing device; receiving user activity tracking data; training a machine learning algorithm, based on the user activity tracking data, to obtain a trained machine learning content recommendation engine; and utilizing the trained machine learning content recommendation engine to identify updated trackable personalized online content based on a plurality of markups to the trackable personalized online content.

Hosting systems configured to generate optimized content based on dynamic tokens and methods of use thereof

A domain name registrar may create a customizable universal purchase path for all customers purchasing a given product. The registrar may create and store modular cross sell components that each has material for at least one product offered for sale by the registrar. Thereafter, the registrar may receive a request for a product from a customer. The registrar may determine a probability the customer will complete the purchase if shown upgrades and/or additional products. If the probability is low (or lower than a predefined threshold) the registrar may complete the purchase before displaying additional upgrades and/or additional products to the customer. This will maximize the conversion rate. Alternatively, if the probability is high (or higher than a predefined threshold) the registrar may create final page markups using the modular cross sell components and collected content and display the final page markups to the customer before completing the purchase with the customer. In preferred embodiments, the final page markups are for the upgrades and/or additional products most likely to be purchased by the customer. This will maximize the amount of each purchase.

Universal purchase path that may be optimized for conversion or revenue

The invention provides for at least three processes for detecting the probability of abusive use of a message account for sending large amounts of unsolicited messages, such as spam, to other message accounts. For example, information provided at registration for a new message account can be processed to determine the likelihood of abusive use of that message account. Also, inbound messages can be processed to determine if the message account that sent the inbound message is abusing the use of that message account. Additionally, outbound messages can be processed to determine if the message account that is attempting to send an outbound message is abusing the use of that message account. Each of these three processes can operate separately or in any combination with each other to further improve the probability that abusive use of a message account will be detected promptly and accurately.

Detection of outbound sending of spam

Detecting bulk fraudulent registration of email accounts

Embodiments are directed towards detecting and reporting use by a domain of a message authentication mechanism, such as DomainKeys (DK), and/or DomainKeys Identified Mail (DKIM), and enabling subsequent blocking of messages based, in part, on its usage. When a message is received by an inbound message server, a message source is determined for the message. In one embodiment, the message source is a domain name associated with the sender of the message. Statistics are recorded about the message, including the message source, whether the message is suspect, includes a forged source identifier, employs DK/DKIM message authentication, and the like. The reports may then be sent to various message sources to enable them to determine the extent of use of DK/DKIM message authentication, and to selectively block, re-direct, or forward the messages based, in part, on the use of DK/DKIM message authentication mechanism.

Blocking of spoofed E-mail

Embodiments are directed towards classifying messages as spam using a two phased approach. The first phase employs a statistical classifier to classify messages based on message content. The second phase targets specific message types to capture dynamic characteristics of the messages and identify spam messages using a token frequency based approach. A client component receives messages and sends them to the statistical classifier, which determines a probability that a message belongs to a particular type of class. The statistical classifier further provides other information about a message, including, a token list, and token thresholds. The message class, token list, and thresholds are provided to the second phase where a number of spam tokens in a given message for a given message class are determined. Based on the threshold, the client component then determines whether the message is spam or non-spam.

Spam filtering based on statistics and token frequency modeling

Embodiments are directed towards detecting and reporting use by a domain of a message authentication mechanism, such as DomainKeys (DK), and/or DomainKeys Identified Mail (DKIM), and enabling subsequent blocking of messages based, in part, on its usage. When a message is received by an inbound message server, a message source is determined for the message. In one embodiment, the message source is a domain name associated with the sender of the message. Statistics are recorded about the message, including the message source, whether the message is suspect, includes a forged source identifier, employs DK/DKIM message authentication, and the like. The reports may ten be sent to various message sources to enable them to determine the extent of use of DK/DKIM message authentication, and to selectively block, re-direct, or forward the messages based, in part, on the use of DK/DKIM message authentication mechanism.

Reporting on spoofed e-mail

The invention relates to determining electronic text communication distributed in bulk is likely solicited. In one step, a first electronic and a second electronic submission are received. It is determined that the first electronic submission is likely solicited. A first portion is extracted from the first electronic submission and a second portion from the second electronic submission. The content of the first electronic submission influences extraction of the first portion, and the content of the second electronic submission influences extraction of the second portion. A first code is determined for the first portion and a second code is determined for the second portion, where the first code is indicative of the first portion and the second code is indicative of the second portion. The first code is compared to the second code. It is determined that the second electronic submission is likely solicited, at least in part, in response to comparing the first code to the second code.

Automated solicited message detection

Growing community of inventors

San Francisco, CA, United States of America

Anirban Kundu