AI Application

AI, here also!

AI. AI is the word we listen to the most in today’s life. AI systems are working for us, with us and around us at different scales. Maybe you have some examples in your mind, whether it be speech recognition or face recognition but here are some more fields that you might  not know of. Let’s get a round-up of some not so known AI applications.

1.   Recommendation Systems:

You must have seen a section named ‘Picked By Us For You’ in various e-commerce sites. This section recommends stuff you might want to buy or you bought previously. Usually this applies on e-commerce sites as well as recommendations of books, movies or series, etc.

Well there are three algorithms being used for such application,

·        Item-Based Recommendation:

For each product, a profile is created including its all characteristics like genres, categories etc. Then the algorithm finds the most similar products to recommend.

·        Collaborative Filtering:

The algorithm finds users who liked the products the same as you and shows recommendations of what the other user liked that you don’t know.

·        Hybrid Approach:

This algorithm combines both the above algorithms. Like the item-based collaborative filtering algorithm uses ratings from similar users to you and recommends products which are liked by similar users. Some of it examples would be youtube, netflix, etc.

For reference, this article by trayolabs can be useful. Click Here.

 2.   Games:

 Here is a pretty well-known application of AI. Lot of AI systems have been developed to play games and be equivalent opponents to human players. 

To apply AI in games, it’s all about the planning and understanding how the environment around you changes and adapts appropriately to that change. The artificial agent has to work on the next best action depending upon the state of the game. The idea of a AI in game is to try various game positions and see how the opponent reacts.

Two different planning categories are used to apply AI in games:

·        Decision Making:

The aim is to control bots and give an illusion that bots are making decisions themselves. Decision trees are used for this purpose. We can create characters that look like they are intelligent by creating and combining decision trees.

·        Pathfinding:1

It is used to move the character from place to place by clicking events by the user. In most of the commercial games, the game environment is assumed as a grid. The whole grid is divided into cells and each cell is associated with a weight (value) which depends upon the environment. A* (A star) and its other algorithm are applied on a grid to find the best possible path.

Most of the gaming companies are using AI as an essential asset. In this article, use of AI in gaming companies and its effect has been described so well.

 3.   Transportation Systems:

There are autonomous cars already in existence as there is heavy use of AI in transportation systems. On the other hand, AI is used to reduce traffic jams by controlling signal lights optimally. As another example, systems in locomotives are installed to reduce the energy consumption and failure rate using machine learning and multi-agent systems. 

Ai systems in cars which detects the obstacles from the path and assures the driver breaks the car on time and much more. AI systems also exist as parking assistance systems to fit your car in congested places.

4.   Logistic & Warehouse Management:

Logistic and warehouse management are too complex issues and it creates many challenges that AI can help to overcome. Amazon is already using warehouses’ robotic shelves for stacking and piling. Read a detailed article about it here.

Coordination and cooperation are the main aspects in such cases. Motion and task planning algorithms as well as multi-agent systems are used in such management systems.

Anticipatory logistics is another big domain for AI. It allows companies to predict spikes and drops in demand based on machine learning techniques. So companies can manage their stacks accordingly.

5.   Design and Architecture:

 Yes! You might be surprised but AI is used in design and architecture too. It is relatively new concept but AI is a magical tool for evidence-based design. It is based on empirical and quantifiable measures. It helps designers and architects to make rational and motivated decisions. 

Conclusion:

I have mentioned a few applications of AI in modern days which are not so known to people. There are many applications I haven’t mentioned. AI is around us and will be more in upcoming years. There are thousands of projects being tested and will be product soon. The AI world is growing and will keep growing.