ML Trends

Newsletter #3 - Mar 17, 2019

Cortana, what is the best time to sow?

It's spring time!

This is the time of the year when folks in the colder regions rejoice shedding those extra layers of clothing and those in the warmer regions rejoice the harvest. It is with this spirit of harvest time that we focus our newsletter on agriculture this edition. AI & ML have a positive impact not just on the sowing and harvesting but also in the commodity marketplace. To know how, keep reading...

Straight from the labs
Cortana, what is the best time to sow?

Microsoft is looking to make a big impact to farming in India with its Farmbeats project. Farmbeats is an initiative of Microsoft to enable data-driven farming as a means to achieve a greater food production in the face of receding water levels and limited arable land.

So far, it has collaborated with International Crop Research Institute for the Semi-Arid Tropics (ICRISAT) and is leveraging new technologies such as AI, Cloud Machine Learning, Satellite Imagery and advanced analytics to help the farmers arrive at the best sowing date for crops, thereby increasing yields by 10-30%. It has also developed a Pest Risk Prediction API in partnership with United Phosphorous Limited (UPL) by leveraging AI and ML to indicate risks of pest attack in advance.

We hope Microsoft becomes the second MS, after MS Swaminathan – the father of green revolution in India, to revolutionise farming in India.

Read the full article here.

Into the realm of business
Can AI create a fair marketplace?
While production is one aspect of the commercial activity of farmers, selling their produce is a gargantuan task in India, especially in non-transparent setups. Farmers are unaware of the price their produce shall fetch at a given quality which renders them incapable of accurately pricing it. At the same time, commodity buyers face challenges with their current buying practices of sampling. They do not have a high level of confidence in the current prevalent assaying processes and are more often than not suspicious about the quality. This leads to them conservatively valuing the goods, thus under-pricing them.

An Indian start up, Intello Labs has leveraged Deep Learning and AI to address this particular issue. Intello Labs’ image-based solutions provide quality monitoring and grading solutions for participants in the agriculture and food industries. With just an image, one can objectively determine the physical composition of a commodity sample within seconds or minutes.

Intello Labs recently developed a solution for cardamom that would identify each individual pod in the sample and classify it by size, color and health. The prototype solution was deployed at the auction house to establish feasibility. Within months of operational activity, the accuracy level of the grading system has reached 95% and is constantly rising, compared to an estimated 70% for manual methods.
Out in the real world
Meet Stan, your valet at the airport.
This bears absolutely no relevance to agriculture, but we couldn’t wait for a week to share these news with you!

France-based Stanley Robotics recently demonstrated Stan, an autonomous robot that can pick up your car from the airport premise and park it for you. The best aspect about Stan is not in the fact that it can park for you, but the possibility that Stan could potentially create 50% more parking space in existing parking lots!

To know more, go through the article on tech crunch here.
We hope you enjoyed this edition of our newsletter. If you've got a couple of minutes to spare, do take this survey and help us get better.

If you found this interesting, you can help us grow by asking your friends, colleagues and enemies to subscribe to our newsletter.
Enjoy the rest of your weekend.

The ML Trends team
Copyright © 2019 mlTrends, All rights reserved.