Inspiration

An idea for predicting when a dispenser will run out of product; An idea for eliminating lag time between alert and refill.

What it does

Sensor Status Prediction And Lag Time Analysis.

How I built it

LSTM network & Random Forest & Neural Net for sensor data prediction. Lag time analysis for dispensers in various sites and locations.

Challenges I ran into

There was a challenge on data processing: trying to figure out the relationship among different relational tables. There was a challenge on building LSTM model: dealing with the structure of LSTM network as well as the structure of input parameters.

Accomplishments that I'm proud of

Figured out relationship among various tables; Successfully built and evaluated LSTM model.

What I learned

General and in-depth knowledge of IoT; knowledge of business analysis on IoT sensor products; ways to deal with time series data; LSTM model.

What's next for Essity Datathon

I am actively looking for internship/full-time job as data analyst, business analyst, machine learning engineer. I am looking forwards to attending more datathons!

Built With

Share this project:

Updates