What we do…

 

T here’s a huge interest in data science at the moment. Businesses understandably want to be a part of it. Very often they assemble the ingredients (the software, the hardware, the team) but then find that progress is slow.

Coppelia is a catalyst in these situations. Rather than endless planning we get things going straight away with the build, learn and build again approach of agile design. Agile and analytics are a natural fit!

Projects might be anything from using machine learning to spot valuable patterns in purchase behaviour to building decision making tools loaded with artificial intelligence. The point is that good solutions tend to be bespoke solutions.

While we build we make sure that in-house teams are heavily involved – trained on the job. We get them excited about the incredible tools that are out there and new ways of doing things. This solves the problem of finding people with the data science skill set. It’s easier to grow your technologists in-house.

The tools are also important. We give our clients a full view of what’s out there, focusing on open source and cloud based solutions. If a client wishes to move from SAS to R we train their analysts not just in R but in the fundamentals of software design so that they build solid, reliable models and tools.

We teach the shared conventions that link technologies together so that soon their team will be coding in python and building models on parallelised platforms. It’s an investment for the long term.

Finally we know how important it is for the rest of the business to understand and get involved with these projects. Visualisation is a powerful tool for this and we emphasize two aspects that are often forgotten: interactivity (even if it’s just the eye exploring detail) and aesthetics: a single beautiful chart telling a compelling story can be more influential than a hundred stakeholder meetings.

 

 

The solutions…

 

We combine an agile design methodology with the latest developments in machine learning and data science to offer you a range of solutions that are…

 

Fast

We’re fast because we know what we are doing. Experience means we pick the right tool for the job first time with no messing around. Combine this with agile and we have a methodology that can deliver results in days not months.

Effective

The best analytics solution in the world is worth nothing unless you can explain it to the business. We are experts in data visualisation and communicating insights. We make sure our projects take root by providing training and mentoring to build up your in-house data science team.

Innovative

An original and innovative response to a problem can change the way your business thinks and give you the edge on competitors. We have the technical and mathematical skills to build tailored solutions from first principles.

7 Day Segmentation

Agile means fast. In just 7 days we can build you a customer segmentation that is ready to use across the business. We work backwards from use cases to ensure relevance. We neither over simplify nor over engineer.

Kick Start Machine Learning

Use data to predict the behaviour of your customers, including acquisition, churn and lifetime value. Applying the latest in machine learning need not require a major investment. We can prove the value in days using open source software.

Rapid Decision Mining

The key question for any analytics project is: how does the available data impact decisions? Success means accessing data quickly and efficiently, understanding business decisions and having the statistical know how to bring the two together even when data is missing or corrupted.

First Principles Build

Some analytical problems can be fiercely difficult, resisting off the shelf solutions and forcing us to revisit the underlying maths. However breakthroughs can transform your business and give you the edge over competitors. We offer bespoke solutions to challenging problems.

Simulated War Games

What will happen to your business when a new competitor enters the marketplace? How might one more recommendation for every ten customers transform your business? We build agent based simulators that demonstrate the consequences of decisions/events in a complex world

Viral Visualisations

A successful visualisation has a life of its own changing the way people see and talk about your business. We balance aesthetics, information content and interactivity to build visualisations that will go viral.

Data Science Incubator

An in house data science capability will change the way your business works. However the skills are hard to find. We advocate growing your own data scientists and offer a range of services to cross train your analysts including tool training, mentoring, coding and stats workshops, project hot housing and hackathons.

Concept Briefing

What exactly is machine learning? What is hadoop and do I need it? Where do I start or what is my next move? What skills should I be looking for? Our concept briefing sessions are for business decision makers who need to quickly get to grips with the key concepts.

Geo-spatial Modelling

Predict your customer’s next purchase from the town and street in which they live. We combine the results of predictive modelling with the latest in mapping technologies to produce tools that will help you understand your customers at a local level.

 

 

Grow your own data scientists…

 

We prioritise skills over tools. There are a lot of people out there building tools but they tend to be about either preprocessing data or prediction and pattern detection for a handful of well defined cases. We love the tools but they don’t address the most difficult problem of how you turn the data into information that can be used in decision making. For that you need skilled analysts wielding tools. Creating the skills is a much harder problem. We have a range of workshops and training programmes that can kickstart your data science team. We can also build bespoke programmes to meet your needs and run problem solving hackathons.

 

Agile for analysts (2 days)

Find out how the agile methodology can revolutionise your analytics projects. Learn to use git and trello to collaborate on ambitious team-led projects

The open source tool set (2 days)

The analyst’s toolbox is now almost limitless. Find out how to use the best of open source technology (e.g. R, python, D3, gephi, mongoDB) to deliver exciting new insights to your business

R for SAS programmers (2 days)

Move from SAS to R as effortlessly as possible . We are experts in both languages and can highlight the differences and similarities. Discover the much vaster world of R and the many possibilities it opens up.

Introduction to data wrangling (2 days)

We all know that the hard part of an analytics project is getting the data in good shape. What is less well known is how tools like R, python, regex and mongoDB can make this simpler for you.

Introduction to machine learning (3 days)

A compressed course that will teach you the essentials of machine learning for prediction, pattern recognition and clustering using R or python

Introduction to AWS and parallelised computing (2 days)

Learn how to speed up machine learning and analytics by deploying it in the cloud over multiple machines. We will be looking at hadoop, hive and spark.

Visualisation: the new tools (2 days)

ALearn how to make beautiful interactive charts and documents in R, d3, gephi and others

Build your own R packages (2 days)

Step gently from coding as an analyst into software design with this introduction to building packages in R.

 

 

The technical part…

 

If you’re thinking of partnering with us or even just interested in digging a little deeper then you’ll probably want some more detail on our areas of expertise. Here it is. Feel free to contact us at simon@coppelia.io for more details.

 

Supervised
A wide range of techniques including regression, neural nets, decision trees, SVM, random forests, nearest neighbour, naive Bayes. Executed in R, Python, Mahout and mllib (Spark)

Unsupervised
Including K-means, hierarchical clustering and latent Dirichlet models. Executed in R, Python, Mahout and mllib (Spark)

Dimension Reduction
Including SVD, PCA, factor analysis and ICA

Feature Selection
Including filter, wrapper and embedded methods

Collaborative Filtering
Item based and user based using Mahout.

Boosting, Bagging and Ensemble Methods

Visualisation in R
Including popular packages ggplot, ggvis and shiny.

D3 and other Javascript libraries

Interactive documents using tangle.js

Network graphs using Gephi and D3

Web based visualisation using plot.ly and infogr.am

Agent based simulation using R and Netlogo

Monte Carlo
Including the simulation of market behaviour using Dirichlet models.

System Dynamics
Modelling feedback loops and other aspects of complex systems using R

Regression
Including more advanced modelling, applicable to real life situations where the assumptions of basic modelling rarely hold. All the major extensions of regression modelling including logistic, log-linear, generalised additive, generalised mixed modelling and robust regressions.

Data Exploration and Dimension Reduction
Including PCA, MDS, Factor Analysis and Self Organising Maps

Bayesian Analysis
Including hierarchical Bayesian models using BUGS and R

Time Series Forecasting
Using Arima, state space exponential smoothing and seasonal decomposition

Computational Methods
Bootstrapping and Monte Carlo Simulation

Optimisation
Including genetic algorithms

Model Deployment and Automation using R and PMML

Languages
R
Python
Javascript (including D3)
Java
SQL
PL-SQL
HiveQL
Matlab
SAS Coding
SPSS Coding
Netlogo
Platforms
AWS
Hadoop
Spark

Version Control
Git
Mercurial

Operating Systems
Linux
Mac OS
Windows
Unix

Databases
Oracle
MS SQL server
Sybase
MongoDB

Machine Learning and Analytics based in London, UK