investorguide.co.za

Home | News | Services | Downloads | Forum | About us | Contact us

Business Analytics

Investorguide is providing a Business Analytics function. We have a team of researchers that specialize in the fields of pattern recognition, machine learning and artificial intelligence (see www.patternrecognition.co.za to learn more about our research).

We specialize in developing solutions for companies that want to employ pattern recognition techniques (e.g. pattern classification and multivariate regression) on business data. 

We have performed pattern classification in various fields such as financial-time-series prediction, image processing, radar image analysis, protein-fold prediction, automatic speech recognition etc. (see our financial time-series results).

What is business analytics?

Business analytics is how organizations gather and interpret data in order to make better business decisions and to optimize business processes. Analytical activities are expanding fast in businesses, government agencies and not-for-profit organizations. 

Machine learning and statistical analysis are integrated with the intuition and knowledge of business analysts in order to discover meaningful and interesting patterns in data.

Analytics are defined as the extensive use of data, statistical and quantitative analysis, explanatory and predictive modeling, and fact-based decision-making. Analytics may be used as input for human decisions; however, in business there are also examples of fully automated decisions that require minimal human intervention. In businesses, analytics (alongside data access and reporting) represents a subset of business intelligence.

How does the process work?

If you have data that you would like to analyse you can contact us at ecomengineering@gmail.com. We will provide you with a quote based on the type of data and analysis that needs to be performed (our quotes are based on hourly rates). A confidentiality agreement will be signed if the quote is accepted in order to ensure the confidentiality of the work and the data. We will then perform the analysis as agreed in the original quote. Specialized software can also be developed in order to analyse new data. 

What type of techniques do Investorguide use to analyse data?

Support vector machine classification and regression, neural network classification and regression, Bayesian classification, Naive Bayes classifiers, Decision Tree classification, k-nearest-neighbour classification and various data measures (see www.patternrecognition.co.za).
 

Examples of successful applications in business analytics

Business analytics has successfully been employed in numerous applications such as finances (portfolio management, forecasting stock price etc.), telecommunications (fraud detection, forecasting network behaviour etc.), marketing (product performance analysis, market segmentation etc.), web analysis (similarity assessment of webpage contents, assessment of user browsing patterns etc.) and others such as insurance claim estimation, hypothesis formulation about illness etc. 

How can my company benefit from business analytics?

A better understanding of your business and business processes can be gained by analysing business data and extracting relevant information from this data. By using advanced statistical and pattern recognition techniques we can identify new and interesting patterns in data which can be used to optimise existing processes and can even be used in making better management decisions.

By making use of business analytics you can gain a significant advantage over your competitors. Expert knowledge can also be encapsulated in software tools that will allow non-experts to make decisions based on expert knowledge (without the inputs/involvement of an expert). 

We can provide specialized software that allows you to perform your own analysis on new data. This software can also be designed to make real-time decisions based on new data. We thus (1) provide insight into the data provided to us and (2) tools to analyse and use similar data in the future. 

Example projects

We have recently used financial time-series data to predict the composite index value of the Johannesburg Stock Exchange (JSE). We have managed to get a relative gain of 37.3 % over the market and an absolute percentage gain of 8.91 for the period of September 2006 to June 2008. Results for this project can be found here.