The Most Interesting Google Patents Ever

The patent reveals Google’s thinking about how best to provide its users with information on the Internet topic that has been hotly debated for years. The patent describes a method that lets users click and drag images around on Google Maps or other maps applications. It also describes a technique that lets users drag images off of one map view and drop them onto another. With patents coming out at a rate of around four to five per day, Google is one of the most patent-rich companies on the planet. This week’s Google patents have us scratching our heads a bit.

Introduction

This is a story about one of the coolest patents ever submitted by Google. The patent itself was so interesting that it inspired us to create the article it describes. Here’s a little backstory on why this patent is special: a few years ago, Google’s Senior Vice President of Search & Video, Amit Singhal, gave a talk called The Evolution of Search. At this talk, he talked about a cool experiment that we had performed back then. One of the projects that Amit talked about involved giving a random person on the street a $5 bill and asking them to go find a friend. Once they found their friend, the person could keep the cash. You may think that Google patents are a boring read. But not so! In a recent patent application, Google has revealed plans to combine the information in all of its existing search engine patents into a single source of knowledge. This would be a valuable resource for anyone looking to get an inside view of Google’s ongoing efforts to keep its search monopoly intact.

Google Patents: Content-Aware Image Enhancement

Google has filed a patent application for content-aware image enhancement that could one day allow Google to identify images and add contextual information to them. The patent, filed last year, describes a system that would recognize an image, then provide contextually relevant information to the user. For example, if you were looking at an image of a woman’s face, Google could provide details such as her age, gender, location, and relationship status. If you looked at a picture of a dog, Google could add details such as breeds, colors, and sizes. While the patent doesn’t provide any specific information on how Google could identify images, it seems like Google already has some of this technology. The company has a patent for using image recognition to

Google Patents: Contextually Aware Language Translation

One of the most interesting Google patents for me as an English speaker was the one titled “Contextually Aware Language Translation”. It seems pretty straightforward at first, but the idea is pretty sophisticated and is certainly worth discussing. I’ll start by explaining the basic premise behind the patent. Suppose that a user is typing something into a search box on a website (as you normally would). But suppose, instead, that there’s an audio-based input method that lets the user speak what they’re writing. You can imagine that a “language translation” system would be able to pick up the user’s speech and translate it to text, or perhaps even voice it back to the user. 

Google Patents: Human-Aware Voice Recognition

The Google patents cover how voice recognition could be improved with the help of a human being. The patent specifically discusses how voice input could be interpreted based on a context of a sentence or phrase rather than a specific word. This means that the technology could be used to detect the speaker’s emotions, tone of voice, and overall feeling of what the user is saying. The system could even go so far as to determine whether a user is trying to tell the truth or lying. 

Google Patents: Information Search Results Based on Voice Input

Google Patents recently published a patent on search results generated based on voice input. The patent application describes a method of providing search results in response to queries submitted by human beings. The patent application describes a system that processes natural language text input to obtain a vector representation of the meaning of the input. Then, the system performs pattern matching and similarity detection on the vector representation, to find entities, phrases, and concepts that match the user query.

Google Patents: Personalized Web Page Recommendations

When it comes to online shopping, personalized web page recommendations are increasingly becoming an area of focus. They allow online retailers to create a user profile of that shopper’s buying habits, interests, and preferences and then provide specific search results for them. To further tailor their results, users can opt-in to see ads alongside their search results. However, some concerns may come up if web users have their data sold off to others, including advertisers, without their consent. The search giant has just applied for a patent for something called “personalized web page recommendations.” According to the patent application, Google would like to provide a “recommendation system for a web browser wherein a web browser automatically receives a recommendation for a web page on a website.” The patent explains that “the recommendation system is based on the user’s preference for a particular content type and the user’s preference for content within a particular category. The system may further include user interaction for modifying the recommendation.”

Conclusion

Google patented a new form of online advertising called “Search by Instant Answer.” This technology would allow users to answer questions online to gain access to specific types of information such as answers to trivia, recipes, and movie reviews. The patent describes the way that Google’s search algorithm works, and it outlines several changes that Google would like to make to the way that search engines work. The biggest change is Google’s proposed new ranking system, which suggests that it would penalize sites that were deemed to be too similar to those that are already ranking high. Google’s suggested new formula was originally designed to provide a better user experience by allowing for the discovery of related pages when users searched for certain terms.