“Being able to do things in real time makes people think differently about the problem.” Andreas Weigend, Director, Social Data Lab

Our customers are industry leaders clamoring for a congruent strategy at the junction of Big Data, Analytics and Business insight. They are leaders of their own industries at the crossroads of endless piles of structured and unstructured data from disparate sources, at the same time overwhelmed with a stream of business challenges that need examination and analytical rigor.

Our customers are not sanguine about their leadership and are always thinking of ways to innovate and stay ahead not merely of the competition, but anticipating the needs of the increasingly demanding 21st century customer who live many lives at once - online, mobile, global, local and blurring the lines between work and play spoilt for choice and hungry for meaning and connection.

Our customers are quant oriented and while they have their hands full with their daily competitive realities – at Helios and Matheson Analytics we enable them to rest a little easier knowing their data is working for them and being worked on by the firm best equipped to nimbly and humbly handle their needs - providing a true customer centric approach giving tribute to technology, humility of data, a willingness of trust along with real world intuitive visualization tools.
Our unique positioning lies in our comfort with the scale and ambitions of enterprise customer questions, discovering ways to motivate and pose the right questions, then integrating using Big Data technology.

We recognize the complexity of sector and company specific business factors as powerful enablers of successful analytics. We understand the crucial need to balance affordability and speed within the parameters of today’s business realties - including the millisecond-by-millisecond collection of data rich customer profiles, importance of preserving privacy, extreme need to safeguard data, various regulatory sensitivities, and other easy to miss risks and organizational sensitivities – and because of these factors we take a careful approach and are cautious in the types of behavioral analysis within our data.
In the past 6 to 8 years we have seen an incredible change in the volume, velocity and variety of data due to emerging technologies in computing power and with the proliferation and use of smart phones - now developed has unleashed the potential of Big Data.

The amount of data in the world is growing fast – outstripping not just the machines, but also our imaginations. In 2013, the amount of stored information in the world is estimated to be around 1,200 exabytes, of which less than 2% is non-digital.

This has caused fundamental shifts in traditional database design - an artifact of small data. Traditional relational databases are challenged to capture, store, search, share, analyze, and visualize data. Big data challenges are handled using a NoSQL(“not only SQL”) database, such as HBase, and may even employ a distributed computing system such as Hadoop. NoSQL databases are typically key-value stores that are non-relational, distributed, horizontally scalable, and schema-free.
While we appreciate the typical approaches of data mining techniques and hypothesis led modeling and equally adept at both, we recognize the potential of either method in posing the right questions and work to integrate models to take advantage of the Ensemble effect. The Ensemble effect suggests that, when joined in an ensemble, predictive models compensate for one another’s limitations, so the Ensemble as a whole is more likely to predict correctly than its component models are, to generate the best outcomes for our customers by cranking up our model’s structural complexity while retaining a critical ingredient a robustness against overlearning. Sampling is no longer necessary as we now have the computing power at our disposal to go thru and normalize numerous sets of data varieties without the need to worry about exactitude.
Data Visualization
For any information to provide valuable insights, it must be interpretable, relevant, and novel. So by having so much unstructured data today, it is critical that the data being analyzed generates interpretable information. Collecting lots of data without the associated metadata — such as what is it, where was it collected, when, how and by whom — reduces the opportunity to play with, interpret, and gain insights from the data.

We design intuitive tools that translate the output of the models into actionable insights. For instance, a clear interface that surfaces the analytic import of compelling real time data and training key employees on how to use the models ensuring front line engagement. This step is crucial to the adoption of predictive analytics and inculcates data oriented thinking in customers.
© 2017 Helios and Matheson. All rights reserved.