ai
Sample Predictive models
Leveraging the tools from Minitab® and PROVALIS Research, along with models we have developed to provide predictions, please find the following samples
Disclaimer for the use of our online Predictive Models
MPR is not:
- Politically Biased, or providing any political position.
- Offering Medical advice.
- Providing Financial advice.
- Providing Professional advice.
Our online predictive models use data we collected from multiple sources on the internet, RSS feeds, and/or manually derived for the purposes of providing examples of how these technologies can be deployed. We provide these models as examples of capabilities we developed, and how they can be leveraged. They are for informational use only, and not professional advice. The actual deployment and implementation of these technologies are client needs specific and require a detailed analysis of data that is to be modeled and the outcome the client is looking to realize. Any utilization of the online models is to be considered a demonstration of these capabilities and should not be considered for any other purpose than an example.
For additional information regarding the use of our website, please see our Terms of Use page.
Data Driven Science, Economics, Public Policy and Business Decision Making
The availability of modern machine learning and text analytics tools enable researchers and business practitioners to raise their predictive analytical capabilities to a new level. The models shown below demonstrate how Wordstat and Minitab products work together to enhance the scientific method.
Theory development and hypothesis testing become much more effective when unstructured data (text) can be mined from journals and social media and then used as economic predictor variables. Model testing using machine learning becomes much more efficient when literally hundreds of predictor variables can be tested for significance at once in a single equation.
The examples shown below give insights as to how today’s powerful computational capabilities are enhancing the philosophical paradigms of the scientific pioneers like Charles Peirce and Karl Popper.
Hover Image in FireFox to expand the view. In Chrome right click, and select open in new tab to expand the view
Predicting Supply-Chain On-Time Delivery
This is an example of predicting the likelihood that a sales order line item will be delivered to a customer on time given historical information available from a typical ERP system.
This would include Bill of Materials, Inventory levels, Routings and work center efficiencies and customer, etc.
In practice the predictive model would be used to identify if the order will be delivered on time based on 1) relevant process factors that impact on-time delivery 2) an understanding of these factors provide delivery dates with greater confidence as commitments to customers.
This example uses a classification model. On-Time or Late. A regression model could also be provided.
Hover Image in FireFox to expand the view. In Chrome right click, and select open in new tab to expand the view
Watch the case study videos below
The videos below review the analysis process and provide context to how this analysis was built. Please feel free to watch these to obtain a better understanding of the theories, data, and analysis that went in to the model.
Discussion 1
Anomaly Detection: Trust Classification in Social Media
Combining Wordstat Text Analytics With Minitab Machine Learning Tools to Predict. A machine learning model that considers known methods of propaganda and deception with identified linguistic forms of deception.
For prediction, enter the values (in words per 10k) to the right, to determine the probability and predicted class of censorship of the subject social media post. The classification of “trust” label is :Low-Medium-High
Hover Image in FireFox to expand the view. In Chrome right click, and select open in new tab to expand the view
Discussion 2
Pattern Recognition: Censorship and Bias Identification in Social Media
For prediction, enter the values (in words per 10k) to the right, to determine the probability and predicted class of censorship of the subject social media post.
Hover Image in FireFox to expand the view. In Chrome right click, and select open in new tab to expand the view
Clients
We work with a wide variety of industries & users
Meta Project Research works with many Businesses in different industries including Industrial Machinery & Components, Life Sciences, Consumer Products, Technology, Oil & Gas, Professional Services, Healthcare, Banking, and Distribution/Sup[ply Chain. We also work with Political Groups and Candidates looking to predict outcomes.
Have a custom project?
We would love to discuss ideas surrounding your next project and help your ideas come to life. Our data scientists and statistical professionals love a challenge.
Browse existing case studies and products
- Tools – To ensure Success
- Knowledge – To obtain Success
- Services – That deliver Success
