GitHub
LinkedIn
Twitter
YouTube
RSS

Electrics Company: Applying Machine Learning Tools to Highlight Anomalous Data

Electrics Company: Applying Machine Learning Tools to Highlight Anomalous Data
CLIENT

Electrics Company

SECTOR

Electronics

The Challenge

Over the last two years, we have been working with a cutting edge electronics company to build advanced algorithms using R and Python. Since they are at the research & design stage of the development process, their data structure is unique and challenging.

Each week they generate thousands of data sets that have a spatial, temporal and experimental component. Their data structure is unique - around 10,000 voltage curves distributed on a disk. This necessitates the use of novel machine learning algorithms, coupled with classical experimental design.

In theory, each disk should be identical. In practice, this isn’t the case. The goal of the project was to develop a set of algorithms and R packages that would identify outliers and deviations from the canonical data. The ultimate goal was to optimise product yield. A novel challenge to this was that any issues detected were then fixed. In fact, our training data set consisted of issues that had now been resolved as they improved their processes.

The Solution

Data from their systems was obtained via APIs. We developed R software to automatically assess the data quality and remove any rogue readings. Second, we created machine learning tools to automatically highlight any potential failures and outliers. Third, we proposed new experimental formulations.

The Results

This was a fast-moving project, where the data type was changing and the data sources were being updated. We developed adaptable software that the client is still using. As the client only has a small data science team, we provided on-site training and regular assistance after the project has finished.

Relevant Case Studies

DiAGRAM: A Shiny app for the National Archives

The National Archives

Jumping Rivers created a feature rich, bespoke Shiny dashboard through a series of development and user testing sprints. Custom widgets were created using React, to give an tailored application. Using a combination of continuous integration and deployment, the dashboard was deployed directly to shinyapps.io via GitHub.

Read more about The National Archives
Scottish Government: Fair Work Data Explorer - Shiny App

Fair Work Data Explorer

Jumping Rivers created a data pipeline and visualisation dashboard for public and policy maker consumption. The aim was to increase transparency and encourage positive change to the workplace environment. The application allows exploration of key performance indicators stratified by a host of protected characteristics.

Read more about Scottish Government
Banking Firm: Code Review & R Package Development

Banking Firm

The client came to Jumping Rivers having already written the code for their problem in VBA. They were trying to evaluate four measurements for agreements with their clients. However, VBA is limited in speed. Jumping Rivers were required to build a bespoke R package to replace and quicken the code.

Read more about Banking Firm
AGR TRACS International: Dashboard Development for Monte Carlo Simulation

AGR TRACS International

At the end of 2017, contacted Jumping Rivers. AGR TRACS International estimates the volumes of oil and gas in subsurface reservoirs. Their work involves combining a set of inputs for each reservoir layer (such as area, thickness, and up to five other inputs) – and then multiplying these inputs together.

Read more about AGR TRACS International
Electrics Company: Applying Machine Learning Tools to Highlight Anomalous Data

Electrics Company

Over the last two years, we have been working with a cutting edge electronics company to build advanced algorithms using R and Python. Since they are at the research & design stage of the development process, their data structure is unique and challenging.

Read more about Electrics Company
Environment Agency: Development & Deployment of 'Water Body Explorer' Shiny App

The Environment Agency

After winning a Water Hub Hackathon, Jumping Rivers were contracted to create a platform for the client using R Shiny, published on RStudio Connect. The platform aggregates raw data from many third party sources both internal and external to the EA, through bespoke APIs, commercial databases, and asset management systems.

Read more about Environment Agency
Financial Institution: Bespoke Report Generation using R Markdown

Financial Institution

The client wanted to assess the viability of R and R Markdown as a reporting tool for creating complex, bespoke documents. We recreated sample reports for them in R Markdown, showcasing that all of their specifications could be met, and provided them with example code and training.

Read more about A Financial Institution
Fujifilm: Shiny Dashboard Creation for Experimental Risk Assessments

Experimental Risk Assessments

Jumping Rivers built a tool for creation of experimental risk assessments via a centralised web application. The dashboard allows for collaborative working during the data entry and assessment formulation, report generation and tracks versions along iteration of the process.

Read more about Fujifilm
NHS Scotland: R Training

NHS Scotland

In the spring of 2018, NHS Scotland expressed a need to move from their existing software, SPSS and SAS, to using R. The difficulty they faced was that there are over two hundred data scientists in NHS Scotland, which made training everyone in the new software a logistical challenge.

Read more about NHS Scotland
Northumbrian Water: Interruption to Supply Risk Mapping using Spatial R Package

NWL Risk Mapping

In spring of 2020 Northumbrian water engaged with Jumping Rivers to build a modelling solution to better understand risks to the consumer within their network in order to provide a better service.

Read more about Northumbrian Water