MACHINE LEARNING
Row Cleaner Evaluation in: Air-Seeders
In the United States, farmers plant hundreds of acres of land at once, right after harvesting. This leaves a thick layer of residue on the ground, and if a row cleaner isn't working properly, it can lead to problems like hairpinning. Hairpinning reduces yield, which in turn means lower profits, and that's simply not acceptable.
To ensure air seeders are performing optimally, we developed a machine learning model with a specialized algorithm to evaluate the performance of row cleaners. The attached video is a sample result from our ML model.
There's more to this project, however I not in a position to show more than the video until we publish the full results. Thank you for understanding
APEGD AI
For meta-analysis, many researchers need to convert scatter plots into tables using online tools like WebPlotDigitizer and PlotDigitizer. This process is time-consuming, as someone has to manually click on each point, calibrate the tool, and then extract the data. Researchers might have to perform this task hundreds of times, depending on the study's requirements.
To address this issue, I, along with my Principal Investigator (PI) at the time, developed an ML tool called APEGD. With APEGD, you can upload an image of a graph, and it will return the data in a neatly organized table, ready for your meta-analysis.
If you want to try out this tool for yourself, please go to https://1data.life/# and click on APEGD AI tab.
TABLEAU DASHBOARDS
Brazilian E-Commerce Sales: Olist Store Insights
This Tableau workbook provides a comprehensive analysis of the Brazilian e-commerce landscape, focusing on data from the Olist Store. Based on the dataset from Kaggle, this dashboard delves into various aspects of the e-commerce experience.
Explore in-depth insights into sales trends, geographical distribution of sellers, top-performing product categories, and delivery performance. Additionally, discover where the majority of customers are located across Brazil.
The interactive dashboard offers highly interconnected data graphs with robust filtering capabilities. Selecting data on one graph dynamically updates and filters the data on the other graphs within the same dashboard, providing a seamless and insightful analytical experience.
Kansas City Crime Stats
This Tableau story offers a comprehensive overview of crime statistics in Kansas City. Explore interactive visualizations that reveal important trends and patterns in crime distribution, frequency, and timing. Delve into insights on regional crime hotspots, seasonal variations, and the impact of different crime categories.
This dynamic dashboard allows in-depth exploration, helping you uncover meaningful insights into Kansas City's crime landscape. Discover how various factors influence crime rates and gain a deeper understanding of the city's safety dynamics.
Note: These statistics may be slightly skewed towards certain regions due to data collection during a period of significant protest activity, resulting in a higher number of arrests.
PANEL DASHBOARD
Brazilian E-Commerce Sales: Olist Store Insights
This is a Panel 3 page Dashboard which provides a comprehensive analysis of the Brazilian e-commerce landscape, focusing on data from the Olist Store. Based on the dataset from Kaggle, this dashboard delves into various aspects of the e-commerce experience but this time in Panel Python Dashboard.
Explore in-depth insights into sales trends, geographical distribution of sellers, top-performing product categories, and delivery performance. Additionally, discover where the majority of customers are located across Brazil.
The interactive dashboard offers highly interconnected data graphs with robust filtering capabilities. Selecting data on one graph dynamically updates and filters the data on the other graphs within the same dashboard, providing a seamless and insightful analytical experience.
The code will be made public soon for you to look at githib.
Note: Literally the same Insights but in Panel 😅😁