The objective of this thesis is to develop a low-cost infrared absorption spectroscope based on linear variable filter (LVF) technology for the automated detection of gases and vapors, and the semi-automated detection of liquids. This instrument represents an alternative to electronic-nose instruments based on cross-selective gas sensor arrays. Instead, the proposed instrument uses the idea of computational “pseudo-sensors”, whereby spectral lines in an analytical instrument are clustered into groups and used as independent variables. We characterize the system on a number of performance metrics, uncovering its detection limits and resolving power. We present calibration methods to estimate the concentration of analytes in a matrix of absorbing species, as well as signal processing techniques for spectral classification. Specifically, we validate the instrument on a mixture calibration problem with simple and complex chemicals, and compare the efficiency of different calibration methods to estimate the concentration of one analyte in the matrix. Moreover, we demonstrate the use of the instrument on two real-world applications in the foodstuffs domain: oil adulteration and trans fatty acid (TFA) detection. The instrument, combined with signal processing techniques, is able to fully discriminate oils, as well as classify margarine and spreads onto high-TFA and low-TFA groups. Despite operating at a low spectral resolution, our results show that the LVF based spectroscope is a promising alternative to traditional analytical techniques for selected niche applications.