BR102015026814A2

BR102015026814A2 - PROCESS OF PH ESTIMATION IN PALMITO

Info

Publication number: BR102015026814A2
Application number: BR102015026814-9A
Authority: BR
Inventors: Barba Bellettini Marcelo; Bellettini Sebastião; Assumpção Fiorda Fernanda; Aparecida Maieves Helayne; Hoffmann Ribani Rosemary
Original assignee: Universidade Federal Do Paraná
Priority date: 2015-10-22
Filing date: 2015-10-22
Publication date: 2017-07-11

Abstract

a presente patente de invenção refere-se ao processo de previsão de ph em palmito, caracterizado pela utilização de fatores agro-meteorológicos como variáveis da equação, oferecendo uma nova ferramenta para padronização de acidez de palmito em relação a segurança alimentar contra a produção da toxina botulínica pelo clostridium botulinum. a combinação de ph e de dados agro-meteorológicos com ferramentas de análise multivariada para a determinação do teor de acidez em palmito é uma abordagem interessante. na presente invenção, a análise de componentes principais (pca) é capaz de discriminar amostras de palmito, em relação a área de cultivo, enquanto a regressão por mínimos quadrados parciais (pls) possibilita o desenvolvimento de um modelo de previsão de ph com pequenos erros associados. os procedimentos de análise de ph em palmito apresentam a necessidade de repetitividade, porém a modelagem matemática possibilita a ausência de preparo da amostra, envolvendo várias vantagens, relacionadas principalmente com a redução do tempo, custo, ausência de resíduos químicos, elaboração de softwares e potencial aplicação em sistemas on-line.The present patent relates to palm heart ph prediction process, characterized by the use of agro-meteorological factors as variables of the equation, offering a new tool for palm heart acidity standardization in relation to food safety against toxin production. botulinum by clostridium botulinum. The combination of ph and agro-meteorological data with multivariate analysis tools for palm heart acidity determination is an interesting approach. In the present invention, principal component analysis (pca) is able to discriminate palm heart samples in relation to cultivation area, while partial least squares (pls) regression enables the development of a small error ph prediction model. associated. palm heart ph analysis procedures present the need for repeatability, but mathematical modeling allows the absence of sample preparation, involving several advantages, mainly related to the reduction of time, cost, absence of chemical residues, software development and potential application in online systems.

Description

“PROCESSO DE PREVISÃO DE pH EM PALMITO” [001] , A presente invenção refere-se ao processo de previsão de pH em palmito, caracterizado pela utilização de fatores agro-meteorológicos como variáveis da equação, oferecendo uma nova ferramenta para padronização de acidez de palmito em relação a segurança alimentar contra a produção da toxina botulínica pelo Clostridium botulinum."Palm pH Prediction Process" [001] The present invention relates to the palm heart pH prediction process, characterized by the use of agro-meteorological factors as equation variables, offering a new tool for standardizing acidity of palmito in relation to food safety against the production of botulinum toxin by Clostridium botulinum.

[002] , O Brasil é o principal produtor e consumidor de palmito como alimento no mundo (GRAEFE et al., 2013; ESPINOSA-PARDO, 2014). Grande parte do palmito consumido é obtido de colheitas ilegais de Euterpe oleracea Mart. e E. edulis Mart., plantas que possuem sua sobrevivência ameaçada pelo cultivo clandestino. Dois produtos são obtidos a partir deste palmito, o seu meristema apical (creme de palmito) colhidos a partir de plantas jovens com cerca de 1 ano de cultivo e seus frutos colhidos a partir de plantas adultas, quando possuem pelo menos 3 anos de cultivo (LETERME et al., 2005; HERNÁNDEZ-UGALDE, MORA-URPÍ; ROCHA, 2011; ROJAS-GARBANZO, et al., 2011).[002], Brazil is the main producer and consumer of palm hearts as food in the world (GRAEFE et al., 2013; ESPINOSA-PARDO, 2014). Much of the palm heart consumed is obtained from illegal harvests of Euterpe oleracea Mart. and E. edulis Mart., plants that have their survival threatened by clandestine cultivation. Two products are obtained from this palm heart, its apical meristem (palm heart cream) harvested from young plants with about 1 year of cultivation and its fruits harvested from adult plants, when they have at least 3 years of cultivation ( LETERME et al., 2005; HERNÁNDEZ-UGALDE, MORA-URPÍ; ROCHA, 2011; ROJAS-GARBANZO, et al., 2011).

[003] , O palmito é definido como um produto comestível, de forma cilíndrica, formato suave, textura macia e sabor levemente doce, extraído a partir da extremidade superior das haste de certas palmeiras. Compreende o meristema apical e um número variável de folhas internas, ainda não totalmente desenvolvidas e interligadas, sendo cercado e protegido pela bainha de folhas adultas exteriores. É rico em aminoácidos, fibras, minerais e vitaminas, apresentando baixo valor calórico (GALDINO & CLEMENTE, 2008).[003] Palm heart is defined as an edible product of cylindrical shape, smooth shape, soft texture and slightly sweet taste, extracted from the upper end of the stem of certain palm trees. It comprises the apical meristem and a variable number of inner leaves, not yet fully developed and interconnected, being surrounded and protected by the outer adult leaf sheath. It is rich in amino acids, fibers, minerals and vitamins, presenting low caloric value (GALDINO & CLEMENTE, 2008).

[004] , Para o processamento de palmito em conserva, é muito importante garantir a segurança alimentar, a fim de se evitar riscos ao consumidor. A matéria-prima não pode ser submetida a tratamento térmico elevado (> 100 °C), já que as suas características sensoriais, principalmente cor e textura, seriam alteradas significativamente (RESENDE & SAGGIN Jr., 2004). O palmito é classificado como alimento de baixa acidez (pH> 4,6) e conservas permitem condições anaeróbicas favoráveis para o desenvolvimento da toxina botulínica. O Clostridium botulinum é um bacilo Gram-positivo, produtor de esporos, frequentemente encontrado em solo, legumes, frutas, fezes humanas e excrementos de animais (DEROSSI et al., 2011). Quando mantido em condições anaeróbicas, produz uma neurotoxina que provoca o botulismo, podendo resultar em morte para o consumidor que ingere a substância. Palmitos são contaminados por esporos de C. botulinum devido ao contato direto ou indireto entre a palma das mãos e o solo durante as operações de campo (PECK et al., 2008). Assim, a estabilidade microbiológica do produto baseia-se na combinação de pH baixo (<4,5) e pasteurização (COSTA et al., 2010). O procedimento de acidez no palmito deve ser considerado uma prática obrigatória, porque de acordo com estudos de investigação atuais e anteriores, uma variação de pH in natura em palmitos é comum (BELLEGARD et al., 2005).[004] For the processing of pickled hearts of palm, it is very important to ensure food safety in order to avoid risks to the consumer. The raw material cannot be subjected to high heat treatment (> 100 ° C), as its sensory characteristics, especially color and texture, would be significantly altered (RESENDE & SAGGIN Jr., 2004). Heart of palm is classified as a low acid food (pH> 4.6) and preserves allow favorable anaerobic conditions for the development of botulinum toxin. Clostridium botulinum is a gram-positive spore-producing bacillus, often found in soil, vegetables, fruits, human feces and animal droppings (DEROSSI et al., 2011). When maintained under anaerobic conditions, it produces a neurotoxin that causes botulism, which can result in death for the consumer who eats the substance. Palm hearts are contaminated by C. botulinum spores due to direct or indirect contact between the palm and the soil during field operations (PECK et al., 2008). Thus, the microbiological stability of the product is based on the combination of low pH (<4.5) and pasteurization (COSTA et al., 2010). The palm heart acidity procedure should be considered a mandatory practice because, according to current and previous research studies, a fresh pH variation in palm hearts is common (BELLEGARD et al., 2005).

[005] , A agrometeorologia é o ramo da ciência que estuda fenômenos físicos atmosféricos e seu impacto sobre o ambiente rural. Um dos desafios da ciência é de prever, com uma antecedência razoável, os resultados das alterações climáticas e as suas possíveis consequências. A cada ano, a Agrometeorologia vem ganhando mais espaço na tomada de decisões operacionais, especialmente nas atividades agrícolas diárias, devido à estreita relação entre clima e solo, como processos de formação do solo, tais como intemperismo de rocha matriz, transporte de partículas e nutrientes, são em grande parte, determinados pelo clima. De acordo com Campos, Bovi e laderoza (1991), híbridos pupunheira cultivadas em diferentes tipos de solo e condições climáticas apresentam maior pH e menor poder tamponante.[005] Agrometeorology is the branch of science that studies atmospheric physical phenomena and their impact on the rural environment. One of the challenges of science is to predict, with reasonable advance, the results of climate change and its possible consequences. Agrometeorology has been gaining more space each year in operational decision-making, especially in daily agricultural activities, due to the close relationship between climate and soil, such as soil formation processes such as matrix rock weathering, particle and nutrient transport. , are largely determined by the weather. According to Campos, Bovi and Laderoza (1991), peach palm hybrids grown in different soil types and climatic conditions have higher pH and lower buffering power.

[006] , Em várias pesquisas científicas, o processo experimental é realizado de forma univariada, utilizando o sistema clássico de uma variável de cada vez. No entanto, por negligenciar a interação entre as variáveis, o resultado obtido não corresponde necessariamente com as condições que resultam em um resultado verdadeiro. Em sistemas químicos, as variáveis tendem a se correlacionar fortemente, interagindo por meio de mecanismos que resultem em efeitos sinérgicos ou antagonistas. Se este fato é ignorado, o processo de experimentação com muitas variáveis se torna inviável e inconsistente (PERALTA-ZAMORA; MORAIS; NAGATA, 2005). Esta avaliação das amostras experimentais é muito melhor do que o seu estudo univariado. Neste contexto, diferentes métodos de análise em combinação com as ferramentas de análise multivariada têm fornecido muitas aplicações relevantes na área de controle de qualidade de alimentos (LI et al., 2007; SINIJA; MISHRA, 2009; FERNÁNDEZ-CABANAS et al., 2011; SHAO et al., 2011; FAN; ROOS, 2015) em que o método de calibração multivariado mais utilizado para a construção de modelos é o de regressão por mínimos quadrados parciais (PLS) (SENA & POPPI, 2004; PIANTAVINI et al., 2014; LUNAetal., 2015; ROPODI et al., 2015).[006] In various scientific researches, the experimental process is performed univariate using the classic system one variable at a time. However, by neglecting the interaction between the variables, the result obtained does not necessarily correspond with the conditions that result in a true result. In chemical systems, variables tend to correlate strongly, interacting through mechanisms that result in synergistic or antagonistic effects. If this fact is ignored, the process of experimentation with many variables becomes unviable and inconsistent (PERALTA-ZAMORA; MORAIS; NAGATA, 2005). This evaluation of the experimental samples is much better than their univariate study. In this context, different analysis methods in combination with multivariate analysis tools have provided many relevant applications in the area of food quality control (LI et al., 2007; SINIJA; MISHRA, 2009; FERNÁNDEZ-CABANAS et al., 2011 ; SHAO et al., 2011; FAN; ROOS, 2015) in which the most used multivariate calibration method for model building is the partial least squares regression (PLS) (SENA & POPPI, 2004; PIANTAVINI et al. , 2014; LUNAetal., 2015; ROPODI et al., 2015).

[007] , Embora, várias revisões e aplicações interessantes tenham sido publicados recentemente (LEI et al., 2015), observa-se que o potencial de aplicação de métodos multivariados não foi adequadamente explorado para as operações de produtos alimentares de controle de qualidade (MELO; ANDREW; FALEIRO, 2015). Estes modelos são uma escolha interessante para estudar características biológicas e químicas, pelo fato das características agro-meteorológicos de um determinado período poderem ser detalhados (MAZUR et al., 2014). De acordo com uma revisão recente da literatura, até o momento, nenhum estudo foi encontrado para utilização de um modelo preditivo multivariado na determinação de pH de palmito ao longo de 12 meses, o que é relevante, considerando que esta análise pode ser inserida na produção industrial, permitindo conclusões importantes sobre o uso de palmito em conserva.[007] Although several reviews and interesting applications have recently been published (LEI et al., 2015), it is noted that the potential application of multivariate methods has not been adequately exploited for quality control food operations ( MELO; ANDREW; FALEIRO, 2015). These models are an interesting choice to study biological and chemical characteristics, because the agro-meteorological characteristics of a given period can be detailed (MAZUR et al., 2014). According to a recent literature review, no studies have been found so far to use a multivariate predictive model for the determination of palm heart pH over 12 months, which is relevant considering that this analysis can be included in the production. allowing important conclusions on the use of pickled palm hearts.

[008] , A combinação de pH e de dados agro-meteorológicos com ferramentas de análise multivariada para a determinação do teor de acidez em palmito é uma abordagem interessante. Na presente invenção, a análise de componentes principais (PCA) é capaz de discriminar as amostras de palmito, em relação a área de cultivo, enquanto a regressão por mínimos quadrados parciais (PLS) possibilita o desenvolvimento de um modelo de previsão de pH com pequenos erros associados de 0,75%. Os procedimentos de análise de pH em palmito apresentam a necessidade de repetitividade, porém a modelagem matemática possibilita a ausência de preparo da amostra, envolvendo várias vantagens, relacionadas principalmente com a redução do tempo, custo, ausência de resíduos químicos, elaboração de softwares e potencial aplicação em sistemas on-line.[008], The combination of pH and agro-meteorological data with multivariate analysis tools to determine palm heart acidity is an interesting approach. In the present invention, principal component analysis (PCA) is able to discriminate palm heart samples in relation to cultivation area, while partial least squares regression (PLS) enables the development of a small pH prediction model. associated errors of 0.75%. Palm heart pH analysis procedures have the need for repeatability, but mathematical modeling allows the absence of sample preparation, involving several advantages, mainly related to the reduction of time, cost, absence of chemical residues, software development and potential. application in online systems.

[009] , Diante do exposto, é apresentado neste relatório o desenvolvimento de uma equação matemática que propõe a utilização da análise de pH e de dados agro-meteorológicos, associados à regressão por mínimos quadrados parciais (PLS), para a determinação do teor de acidez em diferentes partes do palmito contra a produção da toxina botulínica.In the light of the above, this report presents the development of a mathematical equation that proposes the use of pH analysis and agro-meteorological data, associated with partial least squares regression (PLS), to determine the content of acidity in different parts of the palm against the production of botulinum toxin.

[0010] , A Figura 1 apresenta o fluxograma do método de elaboração do processo de previsão de pH em palmito.[0010], Figure 1 shows the flowchart of the method for elaborating the palm heart pH prediction process.

[0011 ]. As seguintes etapas descrevem o método: [0012] , Análise de componentes principais (PCA) das amostras de pH de palmito(1). O software Origem Pro 8.0 (Northampton, MA 01060, EUA) é utilizado para a construção de matrizes. Os dados são processados em Matlab versão 7.1 (Mathworks Inc.) utilizando o pacote PLS-toolbox 1.5 (Eigenvector Research Inc.) para análise de pH. A análise de componentes principais (PCA) é utilizada para reduzir a dimensionalidade do conjunto de dados para algumas variáveis, chamadas de componentes principais (PCs), as quais descrevem a maior variância dos dados analisados. A técnica fornece um resumo das semelhanças e diferenças entre as amostras em relação à variação de pH (LU et al., 2010). A PCA é elaborada para avaliação da variação da acidez do palmito em função dos meses (12 meses), na qual não se utiliza o pré-processamento dos dados devido à alta homogeneidade das amostras em função das variáveis;[0011] The following steps describe the method: Principal Component Analysis (PCA) of palm heart pH samples (1). Source Pro 8.0 software (Northampton, MA 01060, USA) is used for building matrices. Data is processed in Matlab version 7.1 (Mathworks Inc.) using the PLS-toolbox 1.5 package (Eigenvector Research Inc.) for pH analysis. Principal component analysis (PCA) is used to reduce the dataset dimensionality for some variables, called principal components (PC), which describe the largest variance of the analyzed data. The technique provides a summary of similarities and differences between samples with respect to pH variation (LU et al., 2010). PCA is designed to evaluate the variation of palm heart acidity as a function of months (12 months), in which data preprocessing is not used due to the high homogeneity of the samples as a function of the variables;

[0013] , Análise de componentes principais (PCA) dos dados agro-meteorológicos^), para verificação da influência dos fatores agro-meteorológicos em cada mês analisado, pela mesmo método descrito para determinação de PCA em amostras de palmito na qual se utiliza o PCA com pré-processamento (leave-one-out, auto escalamento) devido à alta amplitude entre os dados de amostragem.[0013], Principal Component Analysis (PCA) of agro-meteorological data ^), to verify the influence of agro-meteorological factors in each month analyzed, by the same method described for PCA determination in palm heart samples using PCA with pre-processing (leave-one-out, auto-scaling) due to the high amplitude between sampling data.

[0014] , Calibração e validação de modelo matemático pelo método de regressão por mínimos quadrados parciais (PLS)(3) é utilizado para construir os modelos de predição de pH. Todas as amostras são divididas em dois subgrupos: grupo de calibração (80% dos dados) e de validação interna (20% dos dados). Os modelos PLS são desenvolvidos para correlacionar a variável X (pH) com Y (fatores agro-meteorológicos) na previsão da concentração do composto de interesse dentro do grupo de validação de teste. A validação cruzada leave-one-out consiste em retirar uma amostra do conjunto de calibração e estimar o seu valor previsto com base em um modelo desenvolvido com todas as outras amostras (PIANTAVINI et al., 2014). Este procedimento é indicado por não perder dados (todas as amostras estão envolvidas no desenvolvimento do modelo), sendo mais adequado para uma pequena quantidade de amostras (OLIVEIRA et al., 2014). Em geral, estima-se que o número ótimo de variáveis latentes é aquela que proporciona o menor valor para a soma dos quadrados dos erros de predição (SENA & POPPI, 2004). Durante este processo, a soma dos quadrados do erro residual previsto (PRESS) é calculada para cada componente da PLS (FILHO, 2009; ZOLGHARNEIN et al., 2015). O melhor número possível de fatores PLS é considerado como aquele o qual o PRESS melhora em pelo menos 2% (WISE & GALLAGHER, 2007). O conjunto de previsão (conjunto independente) é definido para validar os resultados do estudo atual (20% dos dados).[0014] Calibration and validation of the mathematical model by the partial least squares regression method (PLS) (3) is used to construct the pH prediction models. All samples are divided into two subgroups: calibration group (80% of data) and internal validation group (20% of data). PLS models are designed to correlate the variable X (pH) with Y (agro-meteorological factors) in predicting the concentration of compound of interest within the test validation group. Leave-one-out cross-validation consists of taking a sample from the calibration set and estimating its predicted value based on a model developed with all other samples (PIANTAVINI et al., 2014). This procedure is indicated for not losing data (all samples are involved in model development) and is best suited for a small amount of samples (OLIVEIRA et al., 2014). In general, it is estimated that the optimal number of latent variables is the one that gives the lowest value for the sum of squares of prediction errors (SENA & POPPI, 2004). During this process, the sum of squares of the predicted residual error (PRESS) is calculated for each PLS component (FILHO, 2009; ZOLGHARNEIN et al., 2015). The best possible number of PLS factors is considered to be the one which PRESS improves by at least 2% (WISE & GALLAGHER, 2007). The prediction set (independent set) is defined to validate the results of the current study (20% of data).

[0015] , Verificação de Outliers(4) para determinação de amostras com alto valor de leverage e resíduos de Student, os quais atribuem efeitos prejudiciais significativos para o modelo, devendo ser removidos a partir dos dados coletados (PEDRO & FERREIRA, 2005). Os valores são detectados utilizando os critérios de leverage e resíduos de Student, de acordo com a equação 3VL/n (em que VL é o número de variáveis latentes e n é o número de amostras), que define a influência de uma dada amostra em um modelo (FRIZON et al., 2015). O limite do residual de Student (± 2,5) indica baixa concentração da amostra (FILHO, 2009).[0015], Outliers Verification (4) for determination of samples with high leverage value and Student residues, which attribute significant detrimental effects to the model and should be removed from the collected data (PEDRO & FERREIRA, 2005). Values are detected using Student's leverage and residual criteria according to the equation 3VL / n (where VL is the number of latent variables and n is the number of samples), which defines the influence of a given sample on a given sample. model (FRIZON et al., 2015). Student residual limit (± 2.5) indicates low sample concentration (FILHO, 2009).

[0016] , Determinação de equação matemática de previsão de pH em palmito(5) pelo gráfico de loadings, com suas respectivas variáveis, expresso sob forma de vetores, de acordo com os passos anteriormente descritos, conforme Figura 2.[0016], Determination of the mathematical equation of palm heart pH prediction (5) by the loadings graph, with their respective variables, expressed as vectors, according to the steps previously described, as shown in Figure 2.

[0017] , Avaliação do desempenho do modelo PLS(6) de previsão de pH, utilizando RMSECV (erro quadrático médio da validação cruzada), erro relativo médio e seus respectivos coeficientes de regressão (R). O coeficiente de determinação (R2) confere a porcentagem de variação presente nos valores dos componentes verdadeiros, o qual é reproduzida na regressão. Normalmente, o menor RMSECV é utilizado para determinar o número ótimo de fatores.[0017], Performance evaluation of the pH prediction model PLS (6) using RMSECV (mean square error of cross-validation), mean relative error and their respective regression coefficients (R). The coefficient of determination (R2) gives the percentage of variation present in the values of the true components, which is reproduced in the regression. Typically, the smallest RMSECV is used to determine the optimal number of factors.

Relação de Figuras: [0018] , Figura 1 - Fluxograma de elaboração de modelo matemático de previsão de pH em palmito.List of Figures: [0018], Figure 1 - Flowchart for the elaboration of a mathematical model of palm heart pH prediction.

[0019] , Figura 2 - Modelo preditivo multivariado de influências agro-meteorológicas em palmito pupunha ao longo de 12 meses.[0019], Figure 2 - Multivariate predictive model of agro-meteorological influences on pupunha palm hearts over 12 months.

Referências Bibliográficas: [0020] , BELLEGARD, C. R. G., RAUPP, D. S., CHAIMSOHN, F. P., & BORSATO, A. V. (2005). Avaliação de procedimentos de acidificação de conservas de palmito foliar de pupunha (Bactris gasipaes). Acta Scientarum, 27(2), 247-254. http://dx.doi. org/10.4025/actasciagron.v27i2.1842.References: BELLEGARD, C.R. G., RAUPP, D.S., CHAIMSOHN, F. P., & BORSATO, A. V. (2005). Evaluation of acidification procedures for canned palm heart palm (Bactris gasipaes) preserves. Acta Scientarum, 27 (2), 247-254. http: //dx.doi. org / 10.4025 / actasciagron.v27i2.1842.

[0021] , CAMPOS, S. D. S., BOVI, M. L. A., IADEROZA, M. (1991). Caracterização do palmito obtido de algumas combinações híbridas entre Açaí e Juçara cultivadas sob diferentes condições. Pesquisa Agropecuária Brasileira, 26(5), 637-646.CAMPOS, S. D. S., BOVI, M. L. A., IADEROZA, M. (1991). Palm heart characterization obtained from some hybrid combinations between Açaí and Juçara cultivated under different conditions. Brazilian Agricultural Research, 26 (5), 637-646.

[0022] . COSTA, M. L., CRUZ, A. G., WALTER, E. Η. M., FARIA, J. A. F„ SANTANA, A. S., & GRANATO, D. (2010). Hearts of palm in conserve: identity and quality aspects and their implications on food safety. International Food Research Journal, 17, 453-459.[0022] COSTA, M.L., CRUZ, A.G., WALTER, E. Η. M., FARIA, J. A. F. SANTANA, A. S., & GRANATO, D. (2010). Hearts of palm in conserve: identity and quality aspects and their implications on food safety. International Food Research Journal, 17, 453-459.

[0023] . DEROSSI, A., FIORE, A. G., DE PILLI, T, & SEVERINI, C. (2011). A Review on Acidifying Treatments for Vegetable Canned Food. Criticai Reviews in Food Science and Nutrition, 5f(February 2015), 955-964. doi: 10.1080/10408398.2010.491163 [0024] . ESPINOSA-PARDO, F. A.; MARTINEZ, J.; MARTINEZ-CORREA, Η. A. Extraction of bioactive compounds from peach palm pulp (Bactris gasipaes) using supercritical CO2. The Journal of Supercritical Fluids, v. 93, 2-6, 2013.[0023] DEROSSI, A., FIORE, A. G., DE PILLI, T, & SEVERINI, C. (2011). A Review on Acidifying Treatments for Vegetable Canned Food. Critical Reviews in Food Science and Nutrition, 5f (February 2015), 955-964. doi: 10.1080 / 10408398.2010.491163 [0024]. ESPINOSA-PARDO, F. A .; MARTINEZ, J .; MARTINEZ-CORREA, Η. A. Extraction of bioactive compounds from peach palm pulp (Bactris gasipaes) using supercritical CO2. The Journal of Supercritical Fluids, v. 93, 2-6, 2013.

[0025] , FAN, F., & ROOS, Y. H. (2015). X-ray diffraction analysis of lactose crystallization in freeze-dried lactose-whey protein Systems. Food Research International, 67, 1-11. http://dx.doi.Org/10.1016/j.foodres.2014.10.023 [0026] . FERNÁNDEZ-CABANÁS, V. M., POLVILLO, O., RODRÍGUEZ-ACUNA, R., BOTELLA, B., & HORCADA, A. (2011). Rapid determination of the fatty acid profile in pork dry-cured sausages by NIR spectroscopy. Food Chemistry, 124(1), 373-378. http://dx.doi.Org/10.1016/j.foodchem.2010.06.031 [0027] , GALDINO, N. O., CLEMENTE, E. (2008). Palmito de pupunha (Bactris gasipaes Kunth.) composição mineral e cinética de enzimas oxidativas. Ciência e Tecnologia de Alimentos, 28(3), 540-544.[0025], FAN, F., & ROOS, Y. H. (2015). X-ray diffraction analysis of lactose crystallization in freeze-dried lactose-whey protein Systems. Food Research International, 67, 1-11. http://dx.doi.Org/10.1016/j.foodres.2014.10.023 [0026]. FERNÁNDEZ-CABANÁS, V. M., POLVILLO, O., RODRIGUEZ-ACUNA, R., BOTELLA, B., & HORCADA, A. (2011). Rapid determination of the fatty acid profile in dry-cured pork sausages by NIR spectroscopy. Food Chemistry, 124 (1), 373-378. http://dx.doi.Org/10.1016/j.foodchem.2010.06.031 [0027], GALDINO, NO., CLEMENTLY, E. (2008). Peach palm (Bactris gasipaes Kunth.) Mineral and kinetic composition of oxidative enzymes. Food Science and Technology, 28 (3), 540-544.

[0028] , GRAEFE, S.; DUFOUR, D.; ZONNEVELD, M.; RODRIGUEZ, F.; GONZALEZ, A. Peach palm (Bactris gasipaes) in tropical Latin America: implications for biodiversity conservation, natural resource management and human nutrition. Biodiversity Conservation, v. 22, 269-300, 2013.[0028], GRAEFE, S .; DUFOUR, D .; ZONNEVELD, M .; RODRIGUEZ, F .; GONZALEZ, A. Peach palm (Bactris gasipaes) in tropical Latin America: implications for biodiversity conservation, natural resource management and human nutrition. Biodiversity Conservation, v. 22, 269-300, 2013.

[0029] , HERNÁNDEZ-UGALDE, J. A.; MORA-URPÍ, J.; ROCHA, O. J.[0029], HERNÁNDEZ-UGALDE, J. A .; MORA-URPÍ, J .; ROCK, O. J.

Genetic relationships among wild and cultivated populations of peach palm (Bactris gasipaes Kunth, Palmae): Evidence for multiple independent domestication events. Genetic Resources and Crop Evolution, v. 58, n. 4, p. 571-583, 2011. http://dx.doi.org/10.1007/s10722-010-9600-6 [0030] , LEI, W., SONG, Y. L, GUO, X. Y., TU, P. F., & JIANG, Y. (2015).Genetic relationships between wild and cultivated populations of peach palm (Bactris gasipaes Kunth, Palmae): Evidence for multiple independent domestication events. Genetic Resources and Crop Evolution, v. 58, no. 4, p. 571-583, 2011. http://dx.doi.org/10.1007/s10722-010-9600-6 [0030], LAW, W., SONG, Y.L., GUO, XY, TU, PF, & JIANG , Y. (2015).

Habitat differentiation and degradation characterization of Cinnamomi Cortex by 1H NMR spectroscopy coupled with multivariate statistical analysis. Food Research International, 67, 155-162. http://dx.doi.Org/10.1016/j.foodres.2014.10.020 [0031] , LETERME, P.; GARCIA, M. F.; LONDONO, A. M.; ROJAS, M. G.; BULDGEN, A.; SOUFFRANT, W. B. Chemical composition and nutritive value of peach palm (Bactris gasipaes Kunth) in rats. Journal of the Science of Food and Agriculture, v. 85, n. 9, p. 1505-1512, 2005. http://dx.doi.Org/10.1002/jsfa.2146 [0032] , LI, X., HE, Y., WU, C., & SUN, D. W. (2007). Nondestructive measurement and fingerprint analysis of soluble solid content of tea soft drink based on Vis/NIR spectroscopy. Journal of Food Engineering, 82(3), 316-323. http://dx.doi.Org/10.1016/j.jfoodeng.2007.02.042 [0033] , LUNA, A. S., DA SILVA, A. P., PINHO, J. S. A., FERRÉ, J., & BOQUÉ, R. (2015). A novel approach to discriminate transgenic from non-transgenic soybean oil using FT-MIR and chemometrics. Food Research International, 67, 206-211. http://dx.doi.Org/10.1016/j.foodres.2014.11.011 [0034] , MAZUR, L., PERALTA-ZAMORA, P. G., DEMCZUK, B., & HOFFMANN, R. (2014). Journal of food composition and analysis application of multivariate calibration and nir spectroscopy for the quantification of methylxanthines in yerba mate (llex paraguariensis). Journal of Food Composition and Analysis, 35(2), 55-60. http://dx.doi.org/10.1016/jjfca.2014.04.005 [0035] . MELO, J., ANDREW, P. W., & FALEIRO, M. L. (2015). Listeria monocytogenes in cheese and the dairy environment remains a food safety challenge: The role of stress responses. Food Research International, 67, 7590. http://dx.doi.org/10.1016/j.foodres.2014.10.031 [0036] . MORA URPI, J. Método practico para germinacion de semillas de pejybaye. Asbana, v. 3, n. 1, p. 14-15, 1979.Habitat differentiation and degradation characterization of Cinnamomi Cortex by 1H NMR spectroscopy coupled with multivariate statistical analysis. Food Research International, 67, 155-162. http://dx.doi.Org/10.1016/j.foodres.2014.10.020 [0031], LETERME, P .; GARCIA, M. F .; LONDONO, A. M .; ROJAS, M. G .; BULDGEN, A .; SOUFFRANT, W. B. Chemical composition and nutritive value of peach palm (Bactris gasipaes Kunth) in rats. Journal of the Science of Food and Agriculture, v. 85, no. 9, p. 1505-1512, 2005. http://dx.doi.Org/10.1002/jsfa.2146 [0032], LI, X., HE, Y., WU, C., & SUN, D. W. (2007). Nondestructive measurement and fingerprint analysis of soluble solid content of tea soft drink based on Vis / NIR spectroscopy. Journal of Food Engineering, 82 (3), 316-323. http://dx.doi.Org/10.1016/j.jfoodeng.2007.02.042 [0033], LUNA, A. S., DA SILVA, A. P., PINE, J. S. A., FERRÉ, J., & BOQUÉ, R. (2015). A novel approach to discriminate transgenic from non-transgenic soybean oil using FT-MIR and chemometrics. Food Research International, 67, 206-211. http://dx.doi.Org/10.1016/j.foodres.2014.11.011 [0034], MAZUR, L., PERALTA-ZAMORA, P. G., DEMCZUK, B., & HOFFMANN, R. (2014). Journal of food composition and analysis of multivariate calibration and spectroscopy for the quantification of methylxanthines in yerba mate (llex paraguariensis). Journal of Food Composition and Analysis, 35 (2), 55-60. http://dx.doi.org/10.1016/jjfca.2014.04.005 [0035]. MELO, J., ANDREW, P. W., & FALEIRO, M. L. (2015). Listeria monocytogenes in cheese and the dairy environment remains a food safety challenge: The role of stress responses. Food Research International, 67, 7590. http://dx.doi.org/10.1016/j.foodres.2014.10.031 [0036]. MORA URPI, J. Practical method for germination of pejybaye seeds. Asbana, v. 3, no. 1, p. 14-15, 1979.

[0037] . PECK, M. W., Goodburn, K. E., Betts, R. P., & Stringer, S. C. (2008). Assessment of the potential for growth and neurotoxin formation by nonproteolytic Clostridium botulinum in short shelf-life commercial foods designed to be stored chilled. Trends in Food Science and Technology, 19(4), 207216. http://dx.doi.org/10.1016/j.tifs.2007.12.006 [0038] . PERALTA-ZAMORA, P., MORAIS, J. L., NAGATA, J. L. D. M. (2005). Por que otimização multivariada ? Revista Engenharia Sanitária e Ambiental, 10(2), 106-110.[0037] PECK, M.W., Goodburn, K.E., Betts, R.P., & Stringer, S.C. (2008). Assessment of the potential for growth and neurotoxin formation by nonproteolytic Clostridium botulinum in short shelf-life commercial foods designed to be stored chilled. Trends in Food Science and Technology, 19 (4), 207216. http://dx.doi.org/10.1016/j.tifs.2007.12.006 [0038]. PERALTA-ZAMORA, P., MORAIS, J.L., NAGATA, J.L.D. M. (2005). Why multivariate optimization? Sanitary and Environmental Engineering Journal, 10 (2), 106-110.

[0039] . PIANTAVINI, M. S., PONTES, F. L. D., CERQUEIRA, L. B., ZAMORA, P. G. P., & PONTAROLO, R. (2014). Simultaneous spectrophotometric determination of pyrantel pamoate and febantel in pharmaceutical preparations using partial least squares regression. Journal of Analytical Chemistry, 69(10), 948-952. http://dx.doi.org/10.1134/S1061934814100104 [0040] . RESENDE, J. M., & SAGGIN JUNIOR, O. J. (2004).[0039] PIANTAVINI, M.S., BRIDGES, F.L.D., CERQUEIRA, L.B., ZAMORA, P.G. P., & PONTAROLO, R. (2014). Simultaneous spectrophotometric determination of pyrantel pamoate and febantel in pharmaceutical preparations using partial least squares regression. Journal of Analytical Chemistry, 69 (10), 948-952. http://dx.doi.org/10.1134/S1061934814100104 [0040]. RESENDE, J. M., & SAGGIN JUNIOR, O. J. (2004).

Processamento de palmito de pupunheira em agroindústria artesanal e uma atividade rentável e ecológica. Sistemas de Produção. (2003). (1st ed.). Rio de Janeiro: Seropédica.Peach palm heart processing in artisanal agroindustry and a profitable and ecological activity. Production systems. (2003). (1st ed.). Rio de Janeiro: Seropedic.

[0041] . ROJAS-GARBANZO, C.; PÉREZ, A. M.; BUSTOS-CARMONA, J.; VAILLANT, F. Identification and quantification of carotenoids by HPLC-DAD during the process of peach palm (Bactris gasipaes H.B.K.) flour. Food Research International, v. 44, n. 7, p. 2377-2384, 2011. http://dx.doi.org/10.1016/j.foodres.2011.02.045 [0042] . ROPODI, A. I., PAVLIDIS, D. E., MOHAREB, F., PANAGOU, E. Z., & NYCHAS, G. J. E. (2015). Multispectral image analysis approach to detect adulteration of beef and pork in raw meats. Food Research International, 67, 12-18. http://dx.doi.org/10.1016/j.foodres.2014.10.032 [0043] . SENA, M. M, POPPI, R. J. (2004). N-way PLS applied to simultaneous spectrophotometric determination of acetylsalicylic acid, paracetamol and caffeine. Journal of Pharmaceutical and Biomedical Analysis, 34(1), 27-34. http://dx.doi.org/10.1016/jJapna.2003.08.011 [0044] . SHAO, Y., CEN, Y., HE, Y., & LIU, F. (2011). Infrared spectroscopy and chemometrics for the starch and protein prediction in irradiated rice. Food Chemistry, 126(4), 1856-1861. http://dx.doi.Org/10.1016/j.foodchem.2010.11.166 [0045] . SINIJA, V. R., & MISHRA, H. N. (2009). FT-NIR spectroscopy for caffeine estimation in instant green tea powder and granules. LWT - Food Science and Technology, 42(5), 998-1002. http://dx.doi.org/10.1016Zj.lwt.2008.12.013 [0046] . FILHO, P. A. C. Rapid determination of sucrose in chocolat mass using near infraredespectrosopy. Analytica Chimica Acta, v. 631, n. 2, p. 216211,2009. http://dx.doi.org/10.10.16/j.aca.2008.10.049 [0047] . FRIZON, N. T.; OLIVEIRA, G. A.; PERUSSELLO, C. A.; PERALTA-ZAMORA, P. G.; CAMLOFSKI, A. M. O.; ROSSA, U. B.; HOFFMANN-RIBANI, R. Determination of total phenolic compounds in yerba mate (Ilex paraguariensis) combining near infrared spectroscopy (NIR) and multivariate analysis. LWT - Food Science and Technology, v. 60, n. 2, p. 795-801, 2015. http://dx.doi.org/10.1016/j.lwt.2014.10.030 [0048] . JIANG, H.; LIU, G.; MEI, C.; YU, S.; XIAO, X.; DING, Y. Rapid determination of pH in solid-state fermentation of wheat straw by FT-NIR spectroscopy and efficient wavelengths selection. Analytical and Bioanalytical Chemistry, v. 404, n. 2, p. 603-611, 2012. http://dx.doi.org/10.1007/s00216-012-6128-y [0049] . LU, X.; WEBB, M.; TALBOTT, M.; VAN EENENNAAM, J.; PALUMBO, A.; LINARES-CASENAVE, J. Distinguishing ovarian maturity of farmed white sturgeon (Acipenser transmontanus) by Fourier transform infrared spectroscopy: a potential tool for caviar production management. Journal of Agricultural and Food Chemistry, v. 58, n. 7, p. 4056-4064, 2010. http://dx.doi.org/10.1021/jf9038502 [0050] . OLIVEIRA, G. A.; BUREAU, S.; RENARD, C. M. G. C.; PEREIRA-NETTO, A. B.; CASTILHOS, F. Comparison of NIRS approach for prediction of internal quality traits in three fruit species. Food Chemistry, v. 143, n. 15, p. 223-230, 2014. http://dx.doi.org/10.1016/j.foodchem.2013.07.122 [0051] . PEDRO, A. M. K.; FERREIRA, M. M. C. Non destructive determination of solids and carotenoids in tomato products by near-infrared spectroscopy and multivariate calibration. Analytical Chemistry, v. 77, n. 8, p. 2505-2511,2005. http://dx.doi.org/10.1021/ac048651r [0052] . PIANTAVINI, M. S.; PONTES, F. L. D.; CERQUEIRA, L. B.; ZAMORA, P. G. P.; PONTAROLO, R. Simultaneous spectrophotometric determination of pyrantel pamoate and febantel in pharmaceutical preparations using partial least squares regression. Journal of Analytical Chemistry, v. 69, n. 10, p. 948-952, 2014. http://dx.doi.org/10.1134/S1061934814100104 [0053] . SENA, M. M.; POPPI, R. J. N-way PLS applied to simultaneous spectrophotometric determination of acetylsalicylic acid, paracetamol and caffeine. Journal of Pharmaceutical and Biomedical Analysis, v. 34, n. 1, p. 27-34, 2004. http://dx.doi.org/10.1016/j.japna.2003.08.011 [0054] . WISE, B. M., GALLAGHER, N. B. PLS Toolbox for use with Matlab, version 4.0, Eigen-vector Technologies. (1st ed ). Iowa: Manson, 2007.[0041] ROJAS-GARBANZO, C .; PÉREZ, A. M .; BUSTOS-CARMONA, J .; VAILLANT, F. Identification and quantification of carotenoids by HPLC-DAD during the process of peach palm (Bactris gasipaes H.B.K.) flour. Food Research International, v. 44, no. 7, p. 2377-2384, 2011. http://dx.doi.org/10.1016/j.foodres.2011.02.045 [0042]. ROPODI, A. I., PAVLIDIS, D. E., MOHAREB, F., PANAGOU, E. Z., & NYCHAS, G. J. E. (2015). Multispectral image analysis approach to detect adulteration of beef and pork in raw meats. Food Research International, 67, 12-18. http://dx.doi.org/10.1016/j.foodres.2014.10.032 [0043]. SENA, M.M., POPPI, R.J. (2004). N-way PLS applied to simultaneous spectrophotometric determination of acetylsalicylic acid, acetaminophen and caffeine. Journal of Pharmaceutical and Biomedical Analysis, 34 (1), 27-34. http://dx.doi.org/10.1016/jJapna.2003.08.011 [0044]. SHAO, Y., CEN, Y., HE, Y., & LIU, F. (2011). Infrared spectroscopy and chemometrics for starch and protein prediction in irradiated rice. Food Chemistry, 126 (4), 1856-1861. http://dx.doi.Org/10.1016/j.foodchem.2010.11.166 [0045]. SINIJA, V. R., & MISHRA, H. N. (2009). FT-NIR spectroscopy for caffeine estimation in instant green tea powder and granules. LWT - Food Science and Technology, 42 (5), 998-1002. http://dx.doi.org/10.1016Zj.lwt.2008.12.013 [0046]. FILHO, P. A. C. Rapid determination of sucrose in chocolat mass using near infraredespectrosopy. Analytica Chimica Acta, v. 631, no. 2, p. 216211,2009. http://dx.doi.org/10.10.16/j.aca.2008.10.049 [0047]. FRIZON, N. T .; OLIVEIRA, G. A .; PERUSSELLO, C. A .; PERALTA-ZAMORA, P. G .; CAMLOFSKI, A. M. O .; ROSSA, U. B .; HOFFMANN-RIBANI, R. Determination of total phenolic compounds in yerba mate (Ilex paraguariensis) combining near infrared spectroscopy (NIR) and multivariate analysis. LWT - Food Science and Technology, v. 60, no. 2, p. 795-801, 2015. http://dx.doi.org/10.1016/j.lwt.2014.10.030 [0048]. JIANG, H .; LIU, G .; MEI, C .; YU, S .; XIAO, X .; DING, Y. Rapid determination of pH in solid-state fermentation of wheat straw by FT-NIR spectroscopy and efficient wavelengths selection. Analytical and Bioanalytical Chemistry, v. 404, no. 2, p. 603-611, 2012. http://dx.doi.org/10.1007/s00216-012-6128-y [0049]. LU, X .; WEBB, M .; TALBOTT, M .; VAN EENENNAAM, J .; PALUMBO, A .; LINARES-CASENAVE, J. Distinguishing ovarian maturity of farmed white sturgeon (Acipenser transmontanus) by Fourier transform infrared spectroscopy: a potential tool for caviar production management. Journal of Agricultural and Food Chemistry, v. 58, no. 7, p. 4056-4064, 2010. http://dx.doi.org/10.1021/jf9038502 [0050]. OLIVEIRA, G. A .; BUREAU, S .; RENARD, C.M. G. C .; PEREIRA-NETTO, A. B .; CASTILHOS, F. Comparison of NIRS approach for prediction of internal quality traits in three fruit species. Food Chemistry, v. 143, no. 15, p. 223-230, 2014. http://dx.doi.org/10.1016/j.foodchem.2013.07.122 [0051]. PEDRO, A. M. K .; FERREIRA, M. M. C. Non destructive determination of solids and carotenoids in tomato products by near-infrared spectroscopy and multivariate calibration. Analytical Chemistry, v. 77, no. 8, p. 2505-2511,2005. http://dx.doi.org/10.1021/ac048651r [0052]. PIANTAVINI, M. S .; PONTES, F. L. D .; CERQUEIRA, L. B .; ZAMORA, P. G. P .; PONTAROLO, R. Simultaneous spectrophotometric determination of pyrantel pamoate and febantel in pharmaceutical preparations using partial least squares regression. Journal of Analytical Chemistry, v. 69, no. 10, p. 948-952, 2014. http://dx.doi.org/10.1134/S1061934814100104 [0053]. SENA, M. M .; POPPI, R. J. N-way PLS applied to simultaneous spectrophotometric determination of acetylsalicylic acid, acetaminophen and caffeine. Journal of Pharmaceutical and Biomedical Analysis, v. 34, no. 1, p. 27-34, 2004. http://dx.doi.org/10.1016/j.japna.2003.08.011 [0054]. WISE, B.M., GALLAGHER, N.B. PLS Toolbox for use with Matlab, version 4.0, Eigen-vector Technologies. (1st ed). Iowa: Manson, 2007.

[0055] . ZOLGHARNEIN, J.; ASANJARANI, N.; AZIMI, G.; GHASEMI, J. Simultaneous spectrophotometric determination of Ga(III) and Tl(III) by using genetic algorithm based on wavelength selection-partial least squares regression. Journal of Analytical Chemistry, v. 70, n. 2, 148-153, 2015. http://dx.doi.org/10.1134/S1061934815020070 REIVINDICAÇÕES[0055] ZOLGHARNEIN, J .; ASANJARANI, N .; AZIMI, G .; GHASEMI, J. Simultaneous spectrophotometric determination of Ga (III) and Tl (III) by using genetic algorithm based on wavelength selection-partial least squares regression. Journal of Analytical Chemistry, v. 70, no. 2, 148-153, 2015. http://dx.doi.org/10.1134/S1061934815020070 CLAIMS

Claims

1. Palm pH prediction process, characterized by the use of agro-meteorological factors as variables of the equation, offering a tool for palm heart acidity standardization in relation to food safety against the production of botulinum toxin by Clostridium botulinum .;

Palm pH prediction process according to claim 1, characterized by defining palm heart as the apical meristem derived from any genus and species of the Arecaceae family;

Palm pH prediction process according to claim 1, characterized by principal component analysis (PCA) of palm heart pH samples in the following steps: a) Use of Origin Pro software (Northampton, MA 01060, USA) ) or any other type of software is used to build arrays. Principal component analysis (PCA) is used to reduce the dataset dimensionality for some variables, called principal components (PCs), which describe the largest variance of the analyzed data; b) Data are processed in Matlab (Mathworks Inc.) or any other statistical software using the PLS-toolbox 1.5 package (Eigenvector Research Inc.) or any pH data analysis software package. This technique provides a summary of similarities and differences between samples in relation to pH variation; c) Principal component analysis (PCA) to reduce the dataset dimensionality for some variables, called principal components (PCs), which describe the largest variance of the analyzed data. The smallest number of major components is used to discriminate the most significant variables in relation to palm heart pH. The PCA is designed to evaluate the variation of palm heart acidity as a function of months, in which data is preprocessed or not depending on palm heart pH homogeneity as a function of the variables;

Palm pH forecasting process according to Claim 1, characterized by principal component analysis (PCA) of agro-meteorological (climate) data, such as atmospheric pressure (hPa), temperature (oC), precipitation rainfall (mm), relative humidity (%); a) Using Source Pro software (Northampton, MA 01060, USA) or any other type of software is used to build arrays. Principal component analysis (PCA) is used to reduce the dataset dimensionality for some variables, called principal components (PCs), which describe the largest variance of the analyzed data; b) Data are processed in Matlab (Mathworks Inc.) or any other statistical software using the PLS-toolbox 1.5 package (Eigenvector Research Inc.) or any software package for analyzing agro-weather data. This technique provides a summary of similarities and differences between months with respect to variation in agro-meteorological factors; c) Principal component analysis (PCA) to reduce the dataset dimensionality for some variables, called principal components (PCs), which describe the largest variance of the analyzed data. The smallest number of major components is used to discriminate the most significant variables in relation to months. The PCA is designed to assess climate change as a function of months, in which data pre-processing (leave-one-out) is used due to the high homogeneity of months as a function of climate variables;

Palm pH prediction process according to Claim 1, characterized by the calibration and validation of the mathematical model by the partial least squares regression method (PLS) for the construction of pH prediction models, in the following steps: a ) Preparation of two subgroups: calibration and internal validation group. Multivariate models are developed to correlate variable X (pH, independent variables) with Y (agro-meteorological factors, dependent variables) in predicting pH concentration within the test validation group. b) Leave-one-out cross-validation to take a sample from the calibration set and estimate its predicted value based on a model developed with all other palm heart pH samples. This procedure is indicated for not losing data (all samples are involved in model development) and is best suited for a small amount of samples. The optimal number of latent variables is the one that gives the smallest value for the sum of squares of prediction errors. During this process, the sum of squares of the predicted residual error (PRESS) is calculated for each multivariate component. The best possible number of PLS factors is considered to be the one which PRESS improves by at least 2%. c) Verification of outliers to determine samples with high leverage value and Student residues to calculate error as a function of mean or standard deviation, which attribute significant detrimental effects to the model and should be removed from the collected data. Leverage is detected according to the equation 3VL / n (where VL is the number of latent variables and n is the number of samples), which defines the influence of a given sample on a model; d) Obtaining a mathematical equation of palm heart pH prediction by the loadings graph, with their respective variables, expressed as vectors, according to the steps previously described; e) Performance evaluation of the pH prediction PLS model using RMSECV (mean square error of cross-validation), mean relative error and their respective regression coefficients (R). The coefficient of determination (R2) gives the percentage of variation present in the values of the true components, which is reproduced in the regression;

Palm pH prediction process according to claim 1, characterized by the use of a mathematical model for palm heart pH prediction, either in the form of an equation or a vector, using statistical software or not, with or without application. in online systems.