Applying Fuzzy Inference Systems in the Extraction of Chia Cake Extract: Predicting the Mass Yield

Abstract

Chia extract has been increasingly used in the food industry since it is rich in bioactive compounds, such as fatty acids, omega-3 fatty, antioxidants, proteins, vitamins, minerals, and dietary fiber. This extract can be obtained by using conventional extraction techniques (e.g., pressure) on chia seeds. Unfortunately, such techniques are insufficient to access all chemical components present in the seeds matrix, producing a by-product named chia cake that is usually discarded. On the other hand, since chia cake contains significant nutraceutical properties, it is still viable and beneficial to perform extractions of chia extract from chia cake. A typical objective of an extraction is to gather a high mass yield of chia (cake) extract. Since the extraction process is complex and expensive (e.g., in terms of laboratory resources), there is an increasing interest in determining the mass yield based on variables of the extraction like temperature, extraction time, and solvent. In this paper, we study the viability of applying traditional fuzzy inference systems (e.g., based on Mamdani's method) and adaptive neurofuzzy inference systems (ANFIS) for this problem. We propose a fuzzy inference architecture that predicts the mass yield of chia cake extract based on temperature, extraction time, and solvent. Our architecture makes use of fuzzy sets and fuzzy rules in the context of fuzzy inference methods. To design them, we create and use a dataset that contains the mass yield of real extractions conducted in the laboratory under different configurations. Hence, it represents another contribution of this paper and serves as the needed foundation to build the proposed architecture. Further, we conduct a performance evaluation to choose the fuzzy inference system that better fits the architecture. Based on our analysis, ANFIS was the best inference method since it delivered the lesser errors and greater correlations between predicted and observed values. We conclude that fuzzy inference systems are powerful tools for the food industry since they can capture the intrinsic imprecise nature of the extraction process, model the existing non-linear relations of the variables, and represent the expert domain knowledge.

Effects of an alternative bleaching method for yerba mate leaves on total phenolic content: evaluation and prediction with soft‐computing approaches

“…This result indicates a high level of relation complexity of drying temperature, AA assays, and TPC that could not be caught by an expert user of the domain. Similar results were reported by other researchers when comparing ANFIS and other methods 17,21 …”

“…Also, multiple LM and ANFIS were employed in evaluation of migration modeling of phthalate from non‐alcoholic beer bottles 19 and oxidative stability of virgin olive oil 20 . Moreover, different configurations of FIS and ANFIS are used to estimate the mass yield of chia cake extract 21,22 …”

“…20 Moreover, different configurations of FIS and ANFIS are used to estimate the mass yield of chia cake extract. 21,22 However, to the best of the authors' knowledge, a detailed analysis of an alternative bleaching process of yerba mate leaves using water immersion and its influence on AA is still missing in the literature. Further, LM, RSM, FIS, and ANFIS are not explored in this context.…”

Fuzzy inference systems for predicting the mass yield in extractions of chia cake extract

“…The described implementation was employed in our research publication in [8]. After analyzing the measures MAE, RMSE, MAPE, and R 2 , we identified ANFIS 1 as the best approach to predicting the mass yield for all the tested solvents.…”

“…In this article, we present an implementation of fuzzy inference systems (FIS) [2][3][4][5][6] and adaptive neuro-fuzzy inference systems (AN-FIS) [7] to predict mass yield of chia cake extract. The knowledge base of these systems uses our real dataset that contains the mass yield from a series of experiments conducted to analyze extractions of chia extract from chia cake [8]. Our real dataset is also included in our implementation, which makes it fully reproducible.…”