Exploring | Extracting | Evaluating | Encoding |
Data Mining | Compounds | Relationships | Nanostructures |
Libraries | Properties | Similarities | Macromolecules |
Localities | Terms & Codes | Algorithms | (Meta)materials |
Exploring localities and
ambiguous terms
A locality is a
neighborhood. In a broader sense, it is a place in space
with vaguely defined boundaries. The term can refer to a
geographic district, but also to a region in
n-dimensional space; representing, say,
material-property data. Like the word “locality,”
diverse “innocent-looking” terms with multiple,
domain-specific meaning are populating the scientific
literature.
We apply computational linguistics methods to disambiguate terms and recognize synonyms. |
Further, we derive domain-specific dictionaries and glossaries for ontology building. Semantic processing of geographic terms teaches us how to automatically pinpoint names to maps—using the Explore Reno-Tahoe entries as a corpus. To associate nanostructure and material terms with synthesis and property information, we apply CurlySMILES as an interface.
Axeleratio Home | |
ChemRange | CurlySMILES |
ThermoML Mining
>> About ThermoML>> Chemical property range
>> Multiple components
>> Solvent effects
Structure Encoding
>> CurlySMILES language>> CurlySMILES project
>> Enantiomers
>> Homopolymers
>> Thin solid films
Curiosity & Learning
>> Chemical trails>> Mushroom hunting
>> Culturomics
>> The Loomis legacy
>> 2-D solid electrolytes