Projects

Innovative Research & Development

At AlgoTactica, creative thinking is central to our business philosophy. We continuously innovate new software and data processing technologies for business analytics and forecasting. Some of our most recent R&D activities are described in the project summaries below.

Forecasting Customer Purchase Patterns via Enhanced Data Metrics

Many researchers have expressed concern that standard Recency, Frequency, and Monetary (RFM) metrics are insufficient. In combination with RFM history, our enhanced data objects also encapsulate time series patterns for 16 additional metrics pertaining to customer behaviour.

Read More

Comparing Customer Sentiment Across Business Sectors

Natural language processing algorithms for sentiment scoring have been used to analyze 43000 online reviews, in order to generate metrics that permit comparisons of customer satisfaction between four distinct business sectors in a large North American city.

Read More

Better Accuracy in Demand Forecasting with LSTM Neural Networks

For some use cases, Long Short-Term Memory (LSTM) neural networks can be a superior forecasting method. In this study, an LSTM model achieved overall RMS error values that were at least 50% lower in comparison to more commonly-used time series regression strategies.

Read More

Probabilistic Forecasting of Customer Purchase Activity

In non-contractual business settings, anticipation of customer visitation and churn activity relies strongly on structured data analytics. However, predicting individual behaviors is extremely difficult when only short transaction histories exist.

Read More

Improving Demand Forecasts for an Electrical Utility

In today’s competitive power markets, electricity is bought and sold as a commodity at market prices. Accordingly, there can be dramatic increases in costs associated with over or under contracting errors which then require the utility to sell or buy power at a loss on the balancing market.

Read More

Deep Learning for Sales Forecasting and Lift Prediction

Sales data often exhibits high variability, and the predictive relationships might be too complex to accurately capture by traditional modeling methods. This poses a significant challenge, because if the model cannot fully learn the data patterns, then it cannot make accurate predictions.

Read More

Improving Sentiment Classification in Natural Language Processing

Sentiment Analysis evaluates a written statement to identify positive, negative, or neutral opinions. One social media channel often monitored for sentiment trends is Twitter. However, Twitter can contain irrelevant statements that might obscure the sentiment signal of interest.

Read More

High-Speed Feature Selection for Design of Big Data Models

Feature selection identifies a set of independent variables containing the most useful information available in order to make a prediction about a dependent variable. However, achieving accurate feature selection with minimum processing time is always a challenge.

Read More

Distributed Software Platform for Wavelet Neural Networks

Wavelet networks are a newer form of data modeling algorithm that integrate neural network concepts with wavelet analysis. Wavelets are special functions that detect very localized changes in data, and can therefore easily adapt when making predictions in a rapidly varying environment.

Read More

Customer Churn Forecasting Using Gradient Boosting Machines

The ability to anticipate customers who might churn is a priority for any business. By implementing accurate churn prediction models as a guide for effective retention programs, customer defections can be mitigated, thus stemming significant revenue loss.

Read More

Using Matrix-Analytic Strategies to Clean Noisy Data

The Internet of Things (IoT) refers to a network of computers, electronic sensors, and other components for intelligent monitoring. A common problem in sensor monitoring applications involves noise-contamination of the signal that the instrument is tasked to measure.

Read More

Feature Engineering for Increased Forecasting Accuracy

When designing regression models, the data features chosen as inputs might sequester relevant information in a way that offers less predictive power to the model than could be achieved if these inputs were subjected to a preprocessing stage.

Read More

Forecasting Asset Failure in the Airline Industry

When replacement pricing and downtime profit losses are considered, repair of a failed equipment asset can be up to 50% more costly than prior maintenance actions that could have averted the failure event altogether. Therefore, data prognostics for preventative maintenance are essential.

Read More

Efficiently Detecting Trends in Customer Demand Time Series

Customer demand time series can embed a long-term slowly-changing non-linear trend that is impossible to observe directly because of spikes and cycles at daily or other intervals. The Discrete Wavelet Transform provides an efficient and reliable method for revealing these trends.

Read More

Improved Price Forecasting via Gaussian Process Regression

Training with short-duration sequences can leave many machine learning models sensitive to localized noise, thus impeding forecasting accuracy. This study demonstrates the exceptional forecasting power of Gaussian Process Regression models when trained on short time-series histories.

Read More
Back to top