In Artificial Intelligence, a model is an abstract representation of a decision process. Its primary goal is to enable the decision process automation, often applied to business. The model can also help understanding the modeled process itself. Machine Learning models are mathematical algorithms that are “trained” using data. Ideally, the model should also explain the reason behind its decision to help understand the decision process.
Predictive models can provide meaningful analytics and insights in topics such as healthcare, trading or customer relationship, and thanks to their anticipation ability, help gain competitive edges. They also prove themselves valuable in Defense & Space use cases.
Descriptive models are an abstract representation of a system. They help better understanding how internal and external events or behaviors interact within the system. It can be used to optimize a workflow and improve active customers’ ROI.
Fed by qualitative and quantitative data, we build decision models help us perceive, organize, and manage the business rules to improve operational performance. Planning, prices, logistics can benefit from decision models.
In predictive modeling, the outcome defines the performed task.
If the outcome is made of continuous values, it is a regression task. The model will then return a numerical value.
If the outcome is made of two or more categories, it is a classification task. The model will then deliver a class. When there are two classes, we talk about binary classification and multi-classification otherwise.
Several algorithms can perform regression and classification tasks to build predictive models: regression algorithms, Bayesian algorithms, kernel algorithms, decision trees, neural networks, and evolutionary algorithms such as ZGP (the core engine of MyDataModels products).
Regression algorithms gather supervised machine learning techniques, where algorithms are trained before being applied to data to create a prediction. They are useful to assess the causal effect of a (or multiple) variable upon another.
Part of supervised machine learning technique as well, decision trees are used to predict a goal or a target based upon a series of questions. It can operate through classification (categories) or regression (numbers).
Time series are used to comprehend the behavior of a given asset over time, and therefore build accurate predictions about its future. It is done by indexing series of data points in time order, whether they are listed or graphically represented.
A new frontier for AI, Small Data represents up to 85% of all the data collected. Small Data algorithms can work on datasets with little history and yet provide meaningful insights through efficient predictive modeling.
ZGP is a unique mathematical expression engine inspired by evolutionary algorithms. It is able to create simple mathematical expressions that are particularly good at predicting or classifying based on small datasets.
Sensitivity analysis allows one to assess the causal effect of one or more variable(s) upon each other. It helps test the robustness of a model and optimize it. It does so by assessing the uncertainty caused by a given variable.
Data visualization displays raw intel through visual representation. It takes reporting to another level. It can be a means for spotting weak signals, thus generating a competitive edge.
Preparing a dataset for its processing can be harsh for non-data-scientists. Format, outliers, missing values are common setbacks. Sometimes, feature engineering can be required. Quaartz takes care of this for you.
Start extracting value from your data
© MyDataModels – All rights reserved | Credits | Terms of use | Privacy and cookies policy
