Patent Issued for System, method, and program product for interactively prompting user decisions (USPTO 11722737): Aimcast IP LLC

2023 AUG 29 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- According to news reporting originating from Alexandria, Virginia, by NewsRx journalists, a patent by the inventors Arazi, Matan (Santa Monica, CA, US), filed on September 13, 2022, was published online on August 8, 2023.

The assignee for this patent, patent number 11722737, is Aimcast IP LLC (Santa Monica, California, United States).

Reporters obtained the following quote from the background information supplied by the inventors: “Content recommendation systems have been an established industry in which extensive technology has been developed by a variety of companies such as Netflix® and Amazon® to provide users with content tailored to their specific interests. For example, a Netflix® subscriber, upon logging into their account may have categories of video on demand content such as “Because You Watched” or “You May Also Like” displayed to them, including specific content selections generated by a content recommendation system. These conventional content recommendation systems rely on extremely large datasets collected from a large number of users and/or subscribers, including content viewing data, content rating data, navigation data related to user navigation through content menus, and other impressions made on the respective website or digital content application interface.

“However, these conventional systems have an inherent technical flaw in that they cannot reliably recommend content that is tailored to a specific user’s interest when there is only a relatively small set of data available for a limited amount of users of the system. For example, an enterprise content recommendation system does not have access to the vast data sets used in conventional data recommendation system such that conventional systems are unable to provide accurate recommendations targeted and tailored to its users in real-time. This lack of data poses technical challenges in generating content tailored specifically to a single user among a relatively small number of users which provide limited data on which to generate recommendations.

“Lifestyle and health management has been an established industry with extensive content in periodicals, books, membership services, and the like. A long felt need in this industry has been to address a fundamental issue with respect to any health management scheme, which is effective assistance in self-monitoring and health and lifestyle management. Certain programs have addressed this issue by introducing oversight and/or peer support. For example, services such as Weight Watchers® include periodic meetings for members of the program to offer guidance, support, and a certain degree of oversight. With technological advancements in personal devices, certain software applications (“apps”) provide for recording user activity, journaling consumption, monitoring health parameters (e.g., heartrate), etc., which may be tied to traditional health management services like Weight Watchers® and other more modern technological solutions like MyFitnessPal® provided by Under Armor® and “Diabetes Prevention Program” provided by Lark™ (http://www.lark.com/dpp-diabetes-prevention-program/) to name a few.

“Current technology relies largely upon self-reporting (e.g., meal and consumption journaling) with retrospective oversight. Such retrospective analysis, which may include some form of reward and punishment scheme, incentivizes inaccurate reporting (either on purpose or by accident) and fails to provide users with effective support at the moments when support is most needed. Indeed, applications today are not technologically capable of providing a real-time notification for an accurate and timely stimulus. Furthermore, the retrospective reporting gives rise to opportunities for users to cheat the system by inaccurate reportage after the fact. Current systems, methods, and program products are unable to evaluate user activity, user preference to provide precisely timed and situationally targeted prompts and/or stimuli for encouraging lifestyle choices and reinforcing health habits at or before a decision point is being made by a user.”

In addition to obtaining background information on this patent, NewsRx editors also obtained the inventors’ summary information for this patent: “In view of the above, it is an object of the present disclosure to provide a technological solution to address the long felt need and technological challenges faced in conventional content recommendation systems in which limited data is available to provide targeted and tailored content to individual users in real time.

“It is also an object of the present disclosure to provide a technological solution to address the long felt need and technological challenges faced in health management services of procuring precisely timed health management directives, such as prompts, rewards, recommendations, challenges or other stimuli to users, such that positive choices are encouraged (and potentially, rewarded) at moments of decision, in contrast with conventional systems that are based on after-the-fact analysis, reward, and punishment. The present disclosure provides for an automated health care system using machine learning and/or heuristic systems that encourages individuals to make everyday choices by detecting situations in real time at or in advance of a decision point in being made by a user in which stimuli are most likely to be the most successful. Another advantage of the present disclosure is that by providing pre-emptive identification of user decision points and real time data capture, both the ability and inclination of users to provide inaccurate reporting is diminished. Collectively, these advantages work in favor of the users themselves as well because it makes it harder to effectively fool the system, method and program product described herein, thereby increasing user compliance and benefits from using the present disclosure.”

There is additional summary information. Please visit full patent to read further.

The claims supplied by the inventors are:

“1. A method comprising: (a) obtaining, by a personal information module of a content recommendation system from a first user device associated with a first user, second raw time-stamped streaming data including a first set of raw time-stamped sensor data associated with a first tracking time period, (b) processing, in real-time by the personal information module of the content recommendation system, the second raw time-stamped streaming data by the steps of: (1) determining, in real-time, a first location data label associated with a location of the first user device at the first time, (2) updating a lifestyle database by storing the second raw time-stamped streaming data labeled with the first location data label; and (3) notifying a situation module of the content recommendation system that there is updated lifestyle information; © upon receiving the update notification, processing, in real-time by the situation module of the content recommendation system, the updated lifestyle information by performing, the steps of: (1) obtaining, by the situation module, the updated lifestyle information including the second raw time-stamped stream data with an associated first location data label and a second plurality of sets of time-stamped sensor data and corresponding location data labels, (2) generating, by the situation module, a first event stream organized by timestamps associated with each respective location data label corresponding to the second plurality of sets of time-stamped sensor data; (3) analyzing, by the situation module, the generated first event stream against event stream information obtained from a user profile database associated with the first user to determine a predicted event expected to occur within a second tracking time period; (4) upon determining the predicted event expected to occur within the second tracking time period, sending, from the situation module to a manager module of the content recommendation system, the generated first event stream as a first query; (d) generating, by a training set module of the content recommendation system, a first content training set wherein the first content training set comprises: i, a first plurality of the previously identified event streams associated with the first user from the user profile database; ii. a first plurality of available content information associated with a plurality of available content selections from an available content database, wherein the available content database comprises available content information for the plurality of available content selections and for each respective available content selection of the plurality of available content selections; iii. a first plurality of content information associated with a plurality of content selections from a content database, wherein the content database comprises content information for the plurality of content selections and for each content selection of the plurality of content selections; and iv. a first plurality of absorption information associated with the plurality of content selections previously viewed by the first user from an absorption database, wherein the absorption database comprises absorption information for a plurality of content selections previously viewed by the first user and for each respective previously viewed content selection of the plurality of content selections; (e) sending, by the training set module, the first content training set to the manager module; (f) upon receipt of the first query, processing, in real-time by the manager module, the first query by the steps of: (1) providing, by the manager module, the first query as a first data input to a first machine-learning algorithm trained by the first content training set to generate as an output first content selection information; (2) generating, by the manager module via the first machine-learning algorithm, the first content selection information, wherein the first content selection information is situationally targeted such that the first content selection information is sent to provide a real-time notification to the first user via the first user device; (g) sending, by the manager module to a collection module of the content recommendation system: i. the first content selection information associated with accessing a first content selection; ii, the first user identification information associated with the first user; iii. the first event stream identification information associated with the first user and the first content selection information; and (h) sending, by the collection module to the first user device, the first content selection information.

“2. The method of claim 1, wherein the first content selection comprises a plurality of available content.

“3. The method of claim 1, wherein the first content selection consists of one available content.

“4. The method of claim 1, wherein the content information further comprises: iii. respective weighting information indicating a respective recommendation determination for a plurality of users of the content recommendation system.

“5. The method of claim 1, wherein a respective content characterizing tag is generated by a tagging module operatively connected to the content module and stored in the content database.

“6. The method of claim 1, wherein the situational targeting of the first content selection is based on the predicted event.

“7. The method of claim 1, wherein the predicted event is predicted in advance by the content recommendation system based on prior event streams.

“8. The method of claim 1, wherein, prior to step (d), the personal information module sends the second raw-time stamped streaming data with its associated first location data label to the situation module.

“9. The method of claim 1, wherein a second plurality of sets of time-stamped sensor data and corresponding location labels includes less than all of the second plurality of sets of time-stamped sensor data within the threshold period of tracking time.

“10. The method of claim 1, wherein the threshold period of tracking time is one of the following: i. 1 calendar day; ii. 12 hours; iii. 6 hours; iv. 1 hour; and v. 30 minutes.

“11. The method of claim 1, wherein the first event stream includes a plurality of event streams, each organized by timestamps associated with its respective location data label.

“12. The method of claim 1, wherein the first event stream is organized further by one or more of the following: i. each event included in the first event stream’s respective time-stamp; and ii. a routine associated with the first user.

“13. The method of claim 1, wherein the second tracking time period is less than 10 minutes.

“14. The method of claim 1, wherein the first content selection information is sent to a second user device associated with a second user.

“15. The method of claim 1, wherein the first content selection information is sent from the manager module to the content module, and then sent from the content module to first user device.

“16. The method of claim 1, wherein the first machine-learning algorithm utilizes a neural network.

“17. The method of claim 1, wherein the first content training set further includes a plurality of previously identified event streams associated with a second user.”

For more information, see this patent: Arazi, Matan. System, method, and program product for interactively prompting user decisions. U.S. Patent Number 11722737, filed September 13, 2022, and published online on August 8, 2023. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(11722737)&db=USPAT&type=ids

(Our reports deliver fact-based news of research and discoveries from around the world.)