Fecha registrada: July 28 2020

The beauty of working in the research and development team, is to be able to look into the future and anticipate needs that our end users did not even realise they required before. Malvern Panalytical is proud to announce our latest version of the HighScore Plus software, version 4.9. Ever since its conceptualisation, our software developers have made improvements such as automated functionalities like converting Bitmap images, faster and improved Rietveld analysis, faster fitting of large data sets, seamless access to Cambridge Structural database from the HighScore platform and more. 

This webinar is presented by our lead software developer, Dr Thomas Deagen, who is based at our supply center in the Netherlands. He will talk about the easy automation functionalities with the latest version 4.9 of HighScore Plus for X-ray diffraction analysis. For instance smart batches, unsupervised and supervised learning. Interested to learn more? Scroll down to register your interest in our series of XRD webinars.

Learn about these new features of Malvern Panalytical's HighScore Plus XRD software which enable easy automation and more accurate analysis

1) “Smart Batches”: Creating an automated analysis is often hampered by a difficult GUI or because the analysis itself is very complicated and requires decisions, loops or other non-linear elements. 
In our latest release we have solved both obstacles at once, by providing a graphical Flowchart alike design +amp;amp; execution Interface, which can contain decision steps as well as any number of loops. 
Such a “Smart Batch” is simply put together by graphically dragging and connecting Action- and Decision-step boxes. We will show how to assemble such a Flowchart automation and how these “Smart Batches” can be used to solve even complex analytical tasks. 

2) “Unsupervised Learning”: Unsupervised learning features were greatly improved too in our latest release, mainly by adding the very popular t-SNE method to cluster (neighborhood) analysis. We will show, based on some advanced XRD data, how this method compares and excels with respect to other in HighScore(Plus) implemented dimensionality reduction methods, like PCA or MMDS. You will see that the t-SNE method makes it a lot easier to spot any (hidden) pattern in your data. 

3) “Supervised Learning”: Also Supervised learning (model building) was greatly improved, here the method of choice for XRD and spectroscopic data clearly is PLSR. We will show how new features like automated optimization of pre-processing’s as well as automated variable selection can be used to easily improve your quantitative models. In addition, the lengthy cross-validation process was sped up by a factor of about 100, by employing multi-threading. All in all, these additions allow to create better, more accurate predictive models in a much shorter time.

Join our free series of webinars: "Better XRD data analysis and interpretation for materials characterization":
- Webinar 1: Introduction to powder X-ray diffraction. More info
- Webinar 2: Studying battery cathode materials using X-ray diffraction More info
- Webinar 3: Expand your powder XRD applications for materials characterization research More info
- Webinar 4: Good vs bad XRD patterns: how to improve your phase analysis. More info
- Webinar 5: Better XRD data quality: importance of good sample preparation. More info
- Webinar 6: Improving your phase search mapping by defining your elemental range: introduction to elemental analysis using X-ray fluorescence. More info
- Webinar 7: Live demo at your desk - latest high performing XRD Benchtop. More info
- Webinar 8: XRD phase quantification tutorial - improve your XRD data analysis.   More info  
- Webinar 9: XRD applications in the minerals industry More info
- Webinar 10: XRD data analysis on HighScore Plus version 4.9 - what's new?  More info 
- Webinar 11: XRD phase quantification tutorial - crystallinity calculation. More info
- Webinar 12: Range of XRD instruments to aid materials characterization research. More info