Insights 2.0
Device: Mac OS
Category: Business
Price: Free, Version: 1.6.1 -> 2.0 (iTunes)
Description:Make Better Decisions. 64-bit Parallel Self-organizing Knowledge Mining: Model. Predict. Simulate. Discover. Associate. Gain new Knowledge and Insights. From Data. Do you know the value of your data? Find unknown relevant relationships in your scientific, business or personal data easily by using outstanding self-organizing knowledge mining methods for forecasting, classification, identification or simulation tasks in various fields. Insights comes with documentation, extra literature, sample data and models such as: • World Oil Price prediction and simulation till 2025; • Monthly Global Temperature, Ozone concentration, or Sun activity predictions till 2017; • Reproductive Toxicity prediction of chemicals; • Housing value or computer system activity prediction to help you getting started with your data. Insights is a professional, yet user friendly tool for building predictive models from noisy data autonomously. Whether applied to sales predictions, financial and resource planning, engineering problems, climate change, health or life sciences related questions, or mining collections of data from government agencies, for example - Insights opens up a wealth of new possibilities to individuals, small business owners, and scientists that were previously available only to large entities that could afford expensive data mining applications. Feature Highlights: • Brings high-performance Personal Knowledge Mining to users with unprecedented ease of model building and deployment - takes full advantage of the computing power of your Mac. • Hides all complex processes of knowledge extraction, model development, dimension reduction, variables selection, noise filtering, and model validation from the user. • Self-organizes linear or nonlinear, static or dynamic regression models and model ensembles - generates the equation that describes the data. • Checks if, and the extent to which, the developed model reflects a valid relationship or if it just models noise - employs advanced validation methods based on higher-dimensional modeling .• Live Prediction Validation technology - for the first time, gives direct information about model stability for the given input values .• Generated analytical models can be used for Status Quo or What-If prediction, analysis, simulation, or optimization problems. • Optionally, it implements models and model ensembles in Excel - model export requires Microsoft Excel Mac 2011 or 2008 Taking observational data that describes a problem, system, or process, the software constructs a working mathematical model. Compatible with data stored in a variety of popular formats (e.g., Microsoft Excel), its AI-powered modeling algorithms allow users to easily extract new and useful knowledge to support decision-making. Start developing predictive models and model ensembles along with a prediction interval from low to high-dimensional noisy data of up to several hundreds of input variables. Insights will do all the hard work for you, autonomously, and will come up with a validated optimal complex analytical model you can apply in the app. Optionally, export your developed models to Excel to further analyze and deploy models in the familiar environment of Excel. Insights is 64-bit parallel software, which optimally takes advantage of all CPUs and CPU cores of your Mac. The app automatically scales to the number of cores available at runtime. No matter which machine you run the app on, you always use its full power. Insights is a complete redevelopment of our former KnowledgeMiner Classic app and is the vital successor of KnowledgeMiner (yX) for Excel.
What's New
- Introducing cost-sensitive modeling for building tailored, purpose-driven classification models from imbalanced data sets. Now easily build tailored, cost-optimized models according to a cost matrix and prevalence you define based on your problem. Directly get dedicated models for your specific application purposes, for example, one model that minimizes false positives and another one for false negatives, both models individually self-organized and optimized for costs and accuracy. This feature is available via in-app purchase and it requires the basic toolset installed. - Two new examples featuring cost-sensitive modeling added: Gene expression of tumor tissue (2000 inputs) and network intrusion detection representing the problem of identification of rare events (1 out of 10,000 or more). - Additional references added for further reading about the backgrounds and concepts of Insights.
Insights