Predictive analytics is concerned with trawling through historical data to find useful patterns which might be used in the future. As such it employs data mining techniques to find the patterns, and once found and verified they are applied via some scoring mechanism, where each new event is scored in some way (e.g. new loan applicants are scored for suitability or not). The data mining platforms compared in this article represent the most common alternatives many organizations will consider. The analysis is high level, and not a feature by feature comparison – which is fairly pointless in our opinion. The five criteria used to compare the products are:
- Capability – the breadth of the offering.
- Integration – how well the analytics environment integrates with data, production applications and management controls.
- Extensibility – very important and a measure of how well a platform can be extended functionally and how well it scales.
- Productivity – the support a platform offers for productive work.
- Value – likely benefits versus costs.
This form of analysis creates some surprises, but you need to look at the full review to see why a particular offering does so well.
Important: If you want to see the full review click on the score. If you want to visit the vendor website click on the name.
4.4 – Revolution Analytics has taken R (the open source analytics platform) and sanitized it for enterprise use. Some technicians may feel it doesn’t need sanitizing, but business and technology managers would probably disagree. In any case it is very hard to fault – which makes the review quite short.
4.3 – IBM Predictive Analytics. Large corporations looking for enterprise wide analytics capability would be foolish not to consider IBM. This behemoth of a supplier has got it all – at a price. You just have to decide whether you want to pay it.
4.2 – Actian provides a complete big data and analytics environment for enterprise analytics needs. What is more the technology is advanced, facilitating analytics tasks that simply are not possible with many other technologies. These are big claims, but Actian has been working quietly in the background to develop and acquire technology that is certainly way ahead of many big data analytics offerings.
4.2 – FICO provides quite unique technology and elegantly combines predictive analytics with prescriptive analytics and business rules management. It’s a formidable combination of capabilities, and it is now available in the FICO Analytic Cloud, so the technology can be accessed by medium size businesses as well as the large corporations that have traditionally used it.
4.2 – SAS will almost certainly address every analytic need your organization could possibly face, and there is a large and skilled pool of SAS professionals around the world. The fly in the ointment is the cost of the technology, and the decision to go with SAS simply boils down to one of perceived value.
4.1 – Angoss provides business oriented data mining technology, avoiding technical complexity for its own sake, and oversimplified products that cannot deliver. This is a good compromise, and many businesses will derive benefits in their customer oriented operations from employing this technology.
4.1 – KNIME is a highly functional data mining platform with a good graphical interface. It is implemented within the Eclipse environment making it extensible and modifiable. It lacks sophisticated big data support, but most things can be accomplished with relevant skills.
4.1 – RapidMiner is an excellent data mining and statistics platform with a large following. It is in no way an end-user tool and requires a good deal of skill to use. With version 6 the product and company became much more commercial, and the recent acquisition of Radoop puts it in the big data league.
4.0 – Salford Systems offers some very capable data mining technology indeed. It excels particularly in ensemble methods, and since these have proved to be some of the most powerful machine learning techniques Salford has won many competitions. Not for the novice, but something different to the usual algorithms that are used in predictive analytics.
4.0 – SAP InfiniteInsight (formerly known as KXEN prior to acquisition by SAP in 2013) addresses a particular set of predictive analytics problems in several well defined markets (typically, but not exclusively retail, financial services and telecoms). The two very significant features of InfiniteInsight are the speed with which predictive models can be built and the reliability of those models. It is not however a general purpose machine learning or data mining toolbox
3.9 – Blue Yonder is perhaps an appropriate name for this supplier. There are some pretty wild claims made concerning automation and its desirability when using predictive technologies, although the underlying technology is novel and sophisticated. Worth a look, but don’t be taken in by the claims of the marketing people.
3.9 – Dotplot provides data mining, statistics, text mining and predictive analytics tools in an integrated, highly graphical cloud based environment. All that is needed to use dotplot is a browser. Resulting models can be integrated with other applications via web services using SOAP and REST protocols. Dig a little deeper and dotplot is actually a much needed graphical front end to R and Weka functions.
3.9 – Oracle has done a surprisingly good job with its predictive analytics platform. It will in the main only be of interest to existing Oracle users, but the in-database analytics does have a broader appeal.
3.9 – STATISTICA from Dell embraces most of the analytics tools many organizations will need, both large and small. One of the most powerful aspects of the product set is the level of integration, with seamless connections between disparate modes. Statistics, machine learning, data mining and text mining are all at the disposal of the user without having to migrate from one environment to another.
3.8 – BigML aims to make predictive analytics easy, and certainly it provides an easy to use drag and drop, point and click interface. Whether predictive analytics will ever be easy is a different matter – there are many potholes even for experienced analysts. In the main BigML uses decision trees to create models, and some ensemble methods in a cloud based environment.
|Alpine Data Labs||4.0||4.5||3.5||4.0||4.0||4.0|
|Oracle Advanced Analytics||4.0||4.0||4.0||4.0||3.5||3.9|