Research:2026-09-29/Journal article/tracing-the-roots-of-illicit-cannabis-a-machine-learning-assisted-atr-ftir-proof-of-concept-study-for-geograph

Research record29 September 2026Journal article

Tracing the roots of illicit Cannabis: A machine learning-assisted ATR-FTIR proof-of-concept study for geographical origin classification

Mahan Singh, Akanksha Sharma, Vishal Sharma

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy · 2026

Tests whether ATR-FTIR spectroscopy combined with machine learning can classify the geographical origin of seized cannabis. A total of 200 female flower-top samples from four distinct regions of North India gave 1000 spectral observations, which were reduced by principal component analysis and used to train four models. An artificial neural network classified best in training and was validated on test data and on unknown data, and the third principal component, linked to holocellulose, lignin, pectin, THCA and THC, contributed most. The abstract does not name the four regions.

DOI

More research from India

• 16 Sep 2026Cannabis in India: Combining Technology and Tradition for Medicinal Use
• 9 Sep 2026Ethnobotanical wealth of the Indian Himalaya: a synthesis of useful species, traditional knowledge and conservation priorities
• 1 Jan 2024Cannabis in Traditional Indian Alchemy
• 1 Dec 2018Cannabis and Social Change in the Indian Himalayas
• 1 Sep 2018Indigenous uses of wild hemp (Cannabis sativa) by the local inhabitants in Manikaran Valley of Himachal Pradesh, North Western Himalaya

Record created 15 October 2026 · maintained by landrace.wiki editors · page history

Published29 Sep 2026

TypeJournal article

VenueSpectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy

Year2026

DOI10.1016/j.saa.2026.128847

CountryIndia

DOI

2026-09-29 2026-10-15 Tracing the roots of illicit Cannabis: A machine learning-assisted ATR-FTIR proof-of-concept study for geographical origin classification Journal article Chemistry Mahan Singh, Akanksha Sharma, Vishal Sharma Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 2026 10.1016/j.saa.2026.128847 https://doi.org/10.1016/j.saa.2026.128847 Tests whether ATR-FTIR spectroscopy combined with machine learning can classify the geographical origin of seized cannabis. A total of 200 female flower-top samples from four distinct regions of North India gave 1000 spectral observations, which were reduced by principal component analysis and used to train four models. An artificial neural network classified best in training and was validated on test data and on unknown data, and the third principal component, linked to holocellulose, lignin, pectin, THCA and THC, contributed most. The abstract does not name the four regions. India