Personalized Recommendation System

Abstract

The invention relates to an AI-based personalized recommendation system used in Over-The-Top (OTT) media services that deliver media directly to audiences over the internet. This system analyzes users' viewing history, social media interactions, and real-time mood to provide content recommendations that enhance the user experience.

Description

1 TARIFF Personalized Recommendation System TECHNICAL AREA 5 The invention relates, in general, to a personalized recommendation system. The invention is particularly useful for Over-the-Counter media, which delivers media directly to audiences via the internet. Top (OTT) media services use users' viewing history and social media activity. by analyzing user media interactions and real-time mood an AI-based platform that offers content recommendations to enhance your experience It is related to the personalized recommendation system. STATE OF THE ART 15 OTT platforms like Netflix, Prime Video, YouTube, Blu TV, and Gain allow users to... These platforms provide users with access to a wide range of media content. They keep track of the amount of time people spend watching certain content in their databases. With current technology, users are conditioned on the types of content they watch, the subject matter, and Platforms that recommend new content to users based on their defined metadata. It exists. However, these platforms fail to understand users' immediate emotions. they can interpret and consequently relate to the incomplete content, or They do not have a system that analyzes momentary moods and makes recommendations. This 25 Therefore, the reason why the person watched that content is unclear. As a result of a study conducted under the known state of the art, TR2023 / 018725 Application number [number] and titled "A CONTENT RECOMMENDATION SYSTEM" was found. The subject of the application is to provide its users with digital audio and video file services, specifically 30. in multimedia content service applications that offer the type of content the user watches It is related to a system that enables recommendations to be made based on the situation. However, here The system described can understand and interpret users' immediate emotions and act accordingly. 2 depending on the content left unfinished, they relate to it or their momentary emotional states. There is no mention of analyzing and making recommendations. As a result, improvements are being made to the recommendation systems, therefore the above... 5 that will eliminate the aforementioned disadvantages and provide solutions to existing systems New organizational structures are needed. THE PURPOSE OF THE INVENTION The present invention meets the aforementioned requirements and has all the disadvantages of 10 with a personalized recommendation system that eliminates and introduces some additional advantages It is related. The main purpose of the invention is to deliver media directly to viewers via the internet. The viewing history of users used in Over-The-Top (OTT) media services, 15 by analyzing social media interactions and real-time mood an AI-based system that provides content recommendations to improve user experience The goal is to provide a personalized recommendation system. One purpose of the invention is to allow users to access publicly available social media profiles and various online platforms. by following music broadcasts, audiobooks, podcast broadcasts, and content that has been left unfinished. The goal is to establish a relevant context and provide content recommendations. Another purpose of the invention is to improve user experience on OTT platforms. The goal is to provide users with a personalized experience. 25 The structural and characteristic features and all the advantages of the invention are given in the figures below. And thanks to the detailed explanation written with references to these figures, it becomes clearer. This will be understood as such. Therefore, the evaluation should also be based on these forms and details. This should be done taking the explanation into consideration. 30 3 BRIEF DESCRIPTION OF THE FIGURES The best way to utilize the advantages of the existing invention, together with its structure and additional elements. For understanding, it should be evaluated together with the figures explained below. is necessary. Figure 1 shows the block diagram view of the personalized recommendation system, which is the subject of the invention. 5 REFERENCE NUMBERS 1. User interface 20. System server 2. Content recommendation engine 10 2.1 Adaptive learning module 2.2 Notification module 3. Sentiment analysis module 40. Content tracking database 4. Content tracking module 15 50. Largest user database 5. Big data tracking module 6. Big data module 7. Internet DETAILED EXPLANATION OF THE INVENTION This detailed description explains the preferred personalized recommendation system that is the subject of the invention. These structures are solely for the purpose of better understanding the subject and have no It is explained in a way that will not create a limiting effect. 25 The invention, shown in the block diagram in Figure 1, connects media directly to the internet. Used in Over-The-Top (OTT) media services that deliver content to viewers via, Users' viewing history, social media interactions, and real-time sentiment. 30 It is an AI-based personalized recommendation system. The invention's subject is a personalized recommendation system that analyzes users' viewing history, half of their data. the content and metadata left behind, the dialogues in the scene he watched, the actors, 4 Scene details such as the music used and objects displayed can be shared publicly through the profile. such as shared texts, clips, songs, liked or commented-on content. social media interactions, music listened to through online platforms, Media content such as podcasts, clips, and real content marked with a timestamp. By analyzing various data such as mood at the time, it takes into account interests and current mood. 5 It offers more accurate content recommendations to users based on their moods. The system that is the subject of the invention, for the purpose of realizing;  By entering the user’s personal information through a user interface (1) logged into a system server (20),  The type of content that the user watches through the system server (20), 10 content they liked, content they left unfinished, while watching the actions it performs such as rewinding, fast-forwarding, and pausing a content tracking module (4) that collects and records the collected data content tracking database (40),  the user's social media accounts or other publications they use 15 content they like, posts they share, and comments they make on various platforms. and by following the people they follow or have unfollowed on social media. a big data tracking module (5) that collects interactions and the collected data a large user database in which it is recorded (50),  In the mentioned content tracking module (4) and big data tracking module (5) 20 By collecting and analyzing data, it identifies the user's browsing history. a big data module (6) and a module that detects instantaneous mood sentiment analysis module (3),  In the aforementioned big data tracking module (5) and sentiment analysis module (3) By evaluating the data presented, the user's tracking history and 25 Content recommendations that generate personalized suggestions based on mood. engine (2) and  The suggestions prepared in the mentioned content recommendation engine (2) are given to the user. a notification module (2.2) and content suggestions based on user behavior The adaptive 30 which updates its engine (2) in real time learning module (2.1) It includes. The working principle of the personalized recommendation system, which is the subject of this invention, is explained below. It's like this: Basically, first, users' viewing history on the platform, unfinished items... Content, stage information, social media likes, publicly available on other platforms. Data is collected from various sources, such as datasets. The collected data is cleaned, They are classified and labeled. For example, a scene left unfinished in a film is categorized as horror 5. This scene suggests that the user doesn't like horror movies. Similarly... If a user can't watch emotional scenes and stops halfway through, it indicates a problem in their relationship. It is possible to consider this. In this way, the user's browsing history and social media... Interests and content preferences are determined by analyzing interactions and unfinished content. This is determined. Accordingly, every time the user logs into the platform, the system will send half a 10-minute message. The remaining content is compared with user data, and artificial intelligence analysis is performed. It reminds by doing so. Past viewing habits and social media interactions. Personalized recommendations are provided based on this data. In an example scenario of a personalized recommendation system based on an invention, a user 15 When logging into the system server (20) for the first time, via the user interface (1) The user's personal data, such as age, gender, and interests, is recorded. The user system Content tracking module for background analysis as you continue using it. (4) Type of content he / she watches, content he / she likes, content he / she leaves unfinished, watching The actions performed during this process, such as rewinding, fast-forwarding, and pausing, are collected. 20 and is recorded in the content tracking database (40). At the same time, the big data tracking module (5) through the user’s social media accounts or other publications he / she uses content they like, posts they share, comments they make, and followers on various platforms. User interactions are collected by tracking individuals they have followed or unfollowed, and It is recorded in the large user database (50). Content tracking module (4) and big data 25 Data collected in the tracking module (5) is transmitted to the big data module (6) and this data The user’s monitoring history is analyzed by the big data tracking module (5) This is determined. For example, by analyzing this data, the user's recent actions can be identified. It can be shown that he watches movies and leaves romantic comedies unfinished. Social Media analysis indicates that the user is interested in comedy and has recently turned 30. It may show that he / she made posts expressing sadness or stress. Content monitoring module (4) and data collected in the big data tracking module (5) are in addition to the big data module (6) The sequence is transmitted to the sentiment analysis module (3). This data is transmitted to the AI-assisted sentiment analysis module. The user's current emotional state is determined by analyzing it in module (3). 6 For example, the user is interested in action and comedy films, romantic films. It turns out he doesn't like comedies and is probably in a bad mood right now. can be placed. It appears in the big data tracking module (5) and the sentiment analysis module (3). The data provided is evaluated by a content recommendation engine (2) and the user Personalized recommendations are provided based on viewing history and mood. 5 For example, the content recommendation engine (2) includes both action and comedy elements, It can suggest a movie that can lift the user's spirits. Content recommendation engine (2) The user is also notified of the incomplete content detected in the content tracking module (4). It can also offer a suggestion to continue with the content. The content tracking module (4) is the same In content that is interrupted at the end, there are scenes that the user might not like. It can skip scenes, inform the user that these scenes were skipped, and conclude the content. can tell the user. Adaptive learning module (2) in the content recommendation engine (2.1) ensures that recommendations are updated and notifications are provided based on user behavior. module (2.2) sends notifications regarding suggestions to the user. Thanks to the personalized recommendation system that the invention offers, the user can simply watch not only their past, but also their social media interactions and current emotions. Based on the user's situation, more accurate and engaging suggestions are provided. This allows the user to spend more time on the system and access more content. By ensuring consumption, both his satisfaction was increased and the system's competitiveness was improved. The advantage has been strengthened.

Claims

7 REQUESTS 1. Over-the-Top (OTT) media, which delivers content directly to viewers via the internet. User viewing history, social media used in media services By analyzing interactions and real-time mood, user 5 an AI-based platform that offers content recommendations to enhance your experience It is a personalized recommendation system, the feature of which is;  By entering the user’s personal information through a user interface (1) logged into a system server (20),  The type of content that the user watches through the system server (20), 10 content they liked, content they left unfinished, while watching the actions it performs such as rewinding, fast-forwarding, and pausing a content tracking module (4) that collects and records the collected data content tracking database (40),  the user's social media accounts or other publications they use 15 content they like, posts they share, and comments they make on various platforms. and by following the people they follow or have unfollowed on social media. a big data tracking module (5) that collects interactions and the collected data a large user database in which it is recorded (50),  In the mentioned content tracking module (4) and big data tracking module (5) 20 By collecting and analyzing data, it identifies the user's browsing history. a big data module (6) and a module that detects instantaneous mood sentiment analysis module (3),  In the aforementioned big data tracking module (5) and sentiment analysis module (3) By evaluating the data presented, the user's tracking history and 25 Content recommendations that generate personalized suggestions based on mood. engine (2) and  The suggestions prepared in the mentioned content recommendation engine (2) are given to the user. a notification module (2.2) and content suggestions based on user behavior The adaptive 30 which updates its engine (2) in real time learning module (2.1) It includes.

Images