b'In a constantly evolving marketplace where new sources and new treatments are regularly coming online, the gemological research being done across the world is paramount to the trade. But the industry has its work cut out for it when it comes to keeping up with it all, with technology improving treatment methods and accelerating the rate of discovery and learning. To help, National Jeweler sifted through studies and our own coverage and consulted experts at gem labs around the world to compile this list of 10 of the most important 1 gemological research developments from the past five years. Machine Learning AdvancesResearchSwiss gem lab SSEF started using machine learning to help it analyze the vast amounts of data produced from chemical analysis of gemstones.I n 2016, SSEF started using themulti-dimensional data, about two years agoof client gemstones the lab has tested since groundbreaking GemTOF instru- SSEF started using a non-linear machinethe 1970s along with the Gbelin Reference ment in its facility, which conductslearning algorithm, which can help identifyStone Collection comprising more than chemical analysis on gemstonesrelevant elemental similarities in the data27,000 gems.through laser ablationa processsets, such as trace elements found in gem-whereby tiny amounts of material are re- stones of different origins. moved from the surface of a gemstone usingGbelin Gem Lab, too, has taken steps a focused laser beam.to advance gemstone analysis through Like other laser ablation systems, it ex- machine learning. tracts chemical information from a few tinyThe lab announced in 2020 it was part-laser-ablated pits. But unlike other systems,nering with Swiss research and develop-GemTOF can capture the presence of allment center CSEM to develop a platform elements simultaneously.that will automate the process of deter-Such testing results in large amounts ofmining gemstone authenticity and origin. trace element data that requires the use ofTogether, Gbelin and CSEM will statistical methods to find relevant patternswork to develop machine learning-based that can be used to help determine a gem- algorithms to evaluate standard gemMachine learning supports SSEFs gemologists stones origin. characteristics using existing data sets as ain origin determination, the lab said. To aid in analyzing the vast amounts ofbase, i.e., the catalog of tens of thousands NATIONAL JEWELER 65'