# Smart Ranking of Emails and Files Based on How You Click > IBM's 1999 patent on automatically sorting lists of items, like emails, by watching which ones you click first and updating a mathematical model of your preferences in the background. - **Patent:** US 6370526 - **Original title:** Self-adaptive method and system for providing a user-preferred ranking order of object sets - **Owner:** International Business Machines Corp - **Granted:** 2002 - **Status:** Public domain (expired) - **Times cited:** 135 - **Field:** software, consumer_electronics, ecommerce ## What it does The system observes the order in which a user accesses a first set of objects, such as opening emails in an inbox, and compares this sequence to an 'access hypothesis'—a baseline expectation of how the user would normally browse them (Claim 1). If the user deviates from this baseline (for example, skipping the top three emails to click a receipt further down), the system adapts a 'preference model' (Claim 1). It does this by applying a 'bonus value' to the features of the early-accessed item (Claim 4) and a 'penalty value' to the skipped items (Claim 9). These updated weights, stored in feature vectors, are then used to calculate preferences and re-rank a second set of incoming items (Claim 1). ## What it does NOT cover - Does not cover ranking systems that rely purely on explicit user feedback, such as star ratings, likes, or manual folder sorting. - Does not cover static ranking systems that do not update their underlying preference models dynamically based on user behavior. - Does not cover systems where the baseline expectation (access hypothesis) is not compared against the actual sequence of user interactions. - Does not cover ranking methods that do not use feature vectors to represent the attributes of the items being sorted. ## The clever bit Instead of just tracking what you click, the system compares your actions to an 'access hypothesis' (like assuming you will read top-to-bottom). By measuring how much you deviate from this baseline, it filters out the bias of screen layout. ## Real-world examples 1. Gmail's Priority Inbox sorting 2. Microsoft Outlook's Focused Inbox 3. Algorithmic sorting of search results in enterprise databases ## Why it matters This patent represents an early foundation for algorithmic feeds and smart inboxes. Long before modern AI-driven social media feeds, IBM patented the core loop of implicit feedback: watching what a user ignores or clicks to silently re-order their view. It directly relates to features like Gmail's Priority Inbox or Outlook's Focused Inbox. ## Frequently asked questions ### What does Smart Ranking of Emails and Files Based on How You Click cover? IBM's 1999 patent on automatically sorting lists of items, like emails, by watching which ones you click first and updating a mathematical model of your preferences in the background. ### Who owns patent US 6370526? International Business Machines Corp owns this patent, granted in 2002. ### When does this patent expire? This patent has expired and is now in the public domain — anyone can use the invention freely. ### What is patent US 6370526 cited by? This patent has been cited by 135 later patents that build on its ideas. ### What problem does this patent solve? This patent represents an early foundation for algorithmic feeds and smart inboxes. Long before modern AI-driven social media feeds, IBM patented the core loop of implicit feedback: watching what a user ignores or clicks to silently re-order their view. It directly relates to features like Gmail's Priority Inbox or Outlook's Focused Inbox. ### What does this patent NOT cover? Does not cover ranking systems that rely purely on explicit user feedback, such as star ratings, likes, or manual folder sorting. **Full plain-English explainer:** https://patentbrief.org/patent/us/6370526/google-adwords-pay-per-click **Original patent:** https://patents.google.com/patent/US6370526 --- _Source: PatentBrief — https://patentbrief.org. Patent facts are from public records; the plain-English explanation is PatentBrief's._ ## Related patents Semantically similar inventions in the PatentBrief corpus: - [How Facebook Ranks Search Results Based on Your Friends' Activity](https://patentbrief.org/patent/us/8914392/facebook-events) — A method for ranking search results by prioritizing links that your social network friends have clicked on previously. - [How Google Uses User Feedback to Rank Websites by Quality](https://patentbrief.org/patent/us/8615514/evaluating-website-properties-by-partitioning-user-feedback) — A method for Google to judge a website's quality by looking at how users interact with its search results across different categories of search queries. - [How Browsers Automatically Choose Your Start Page Based on History](https://patentbrief.org/patent/us/6266060/windows-desktop) — An IBM patent from 1998 that describes how a web browser can automatically pick which website to load first based on your past browsing habits. - [How a Recommendation System Learns What You Really Want to Watch](https://patentbrief.org/patent/us/12118030/dynamic-feedback-in-a-recommendation-system) — This patent describes a media recommendation system that creates a unique user profile, uses it to pick a starting point, and then dynamically adjusts a list of suggestions based on what you've already watched and your real-time feedback. - [How Computers Automatically Organize and Search Photos Using Contextual Data](https://patentbrief.org/patent/us/8862582/amazon-redshift) — A system for indexing images by attaching descriptive data to objects within them and adjusting search rankings based on how often a user searches for those specific items.