Freshness Classification of Horse Mackerels with E-Nose System Using Hybrid Binary Decision Tree Structure

Abstract

The aim of this study is to test the freshness of horse mackerels by using a low cost electronic nose system composed of eight different metal oxide sensors. The process of freshness evaluation covers a scala of seven different classes corresponding to 1, 3, 5, 7, 9, 11, and 13 storage days. These seven classes are categorized according to six different classifiers in the proposed binary decision tree structure. Classifiers at each particular node of the tree are individually trained with the training dataset. To increase success in determining the level of fish freshness, one of the k-Nearest Neighbors (k-NN), Support Vector Machines (SVM), Linear Discriminant Analysis (LDA) and Bayes methods is selected for every classifier and the feature spaces change in every node. The significance of this study among the others in the literature is that this proposed decision tree structure has never been applied to determine fish freshness before. Because the freshness of fish is observed under actual market storage conditions, the classification is more difficult. The results show that the electronic nose designed with the proposed decision tree structure is able to determine the freshness of horse mackerels with 85.71% accuracy for the test data obtained one year after the training process. Also, the performances of the proposed methods are compared against conventional methods such as Bayes, k-NN, and LDA.

Recent Progress in Electronic Noses for Fermented Foods and Beverages Applications

“…In terms of food business competition, they have been used to analyze aromas and compare them with competitor products [103,104], evaluate the impact of changes in the production process and components that affect organoleptic characteristics [105,106] and compare different food formulations [84,107]. Moreover, E-nose systems have showed high performance in identifying the quality of many products, including wine [108], beer [109], coffee [110], carbonated drinks [111], dairy products [112,113], pork [114], beef [115,116], chicken [117], fish [118][119][120] and shrimp [121,122]. However, the sensors in E-nose systems may have a drift effect.…”

“…The variety of E-nose applications to detect odors from food and beverage products caused by alcoholic fermentation is summarized in Table 1. [112,113], pork [114], beef [115,116], chicken [117], fish [118][119][120] and shrimp [121,122]. However, the sensors in E-nose systems may have a drift effect.…”

Determination of Chopped Fruits Freshness with High Accuracy by Using Electronic Nose

“…The quality or aroma of many foods such as tomatoes [12], peaches [13] and beverages such as fruit juice[14], wine [15] and tea [16] have been determined with the e-nose in food industry. Moreover, there are many e-nose studies available on the open literature to determine the shelf life of milk [17] and freshness of food such as fish [18], peach [19], egg [20], etc. In this study; daily changes of the smell of chopped melon, peach, banana and uncut strawberry have been observed with electronic nose.…”

A Novel Multiway Splits Decision Tree for Multiple Types of Data

“…Despite the great success of deep neural network (DNN) model in image processing, speech recognition, and other fields in recent years, decision trees have competitive performance compared to DNN scheme, such as the advantage of interpretability, less parameters, and good robustness to noise, and can be applied to large-scale data sets with less computational cost. erefore, the decision tree is still one of the hotspots in the field of machine learning today [1][2][3]. e research mainly focused on the construction method of decision trees, split criterion [4], decision trees ensemble [5,6], mixing with other learners [7][8][9], decision trees for semisupervised learning [10], and so on.…”

“…To solve the problem of parallel decision trees, some researchers introduced oblique decision trees. In such oblique decision trees, the nonleaf node tests the linear combination of features, i.e., p i�1 a l x l ≤ θ, (2) where a l represents the coefficient for the lth feature, θ is the threshold, and p is the number of features. In Figure 1, the instances of the two classes can be completely separated by one oblique split.…”