Classification

3. Classification Models (vision models/ llm text classification/ sentiment analysis)

Image input and structured data output.

• A farming robot wants to selectively target weeds instead of crops. It uses image classification to mark which plant is what type and uses lasers to burn the weeds.
• A fishing vessel uses nets to catch fish. It only wants to collect specific species and sizes it uses classification and robots to release the unnecessary species.
• A recycling plant wants to remove certain materials from a conveyer belt to separate material types. A model scans it and a picking robot removes the materials.
• A potato farmer can't sell green potatoes so he has a machine that uses vision to determine which potatoes are bad and remove them from the processing line.
• A logging company has 100,000 acres of land that they need to survey to check for growth patterns, determine harvesting schedules, and view new lands for acquisition. They use satellite imagery and train an ai model to produce estimations for timber types and amount. A model to detect pest and disease occuring in the forest.
• A cancer clinic is underfunded. They want to provide service to as many patients as possible. It currently takes 1 doctor 4 hours to analyze the imaging and charts of one breast cancer patient. The accuracy rate is 95%. By employing a classification model which takes the images and attributes the classification is instant and the accuracy actually improve to 98%.

100 billion$