Parameters | Percentage (%) |
Moisture | 87.83 ± 0.04 |
Fat | 4.57 ± 0.04 |
Protein | 3.25 ± 0.01 |
Solid not fat (SNF) | 7.57 ± 0.04 |
Total solid (TS) | 12.15 ± 0.03 |
Ash | 0.69 ± 0.01 |
Titrable acidity | 0.13 ± 0.001 |
In the realm of machine learning, where data-driven insights guide decision-making, addressing the challenges posed by class imbalance in datasets has emerged as a crucial concern. The effectiveness of classification algorithms hinges not only on their intrinsic capabilities but also on their adaptability to uneven class distributions, a common issue encountered across diverse domains. This study delves into the intricate interplay between varying class imbalance levels and the performance of ten distinct classification models, unravelling the critical impact of this imbalance on the landscape of predictive analytics. Results showed that random forest (RF) and decision tree (DT) models outperformed others, exhibiting robustness to class imbalance. Logistic regression (LR), stochastic gradient descent classifier (SGDC) and naïve Bayes (NB) models struggled with imbalanced datasets. Adaptive boosting (ADA), gradient boosting (GB), extreme gradient boosting (XGB), light gradient boosting machine (LGBM), and k-nearest neighbour (kNN) models improved with balanced data. Adaptive synthetic sampling (ADASYN) yielded more reliable predictions than the under-sampling (UNDER) technique. This study provides insights for practitioners and researchers dealing with imbalanced datasets, guiding model selection and data balancing techniques. RF and DT models demonstrate superior performance, while LR, SGDC and NB models have limitations. By leveraging the strengths of RF and DT models and addressing class imbalance, classification performance in imbalanced datasets can be enhanced. This study enriches credit risk modelling literature by revealing how class imbalance impacts default probability estimation. The research deepens our understanding of class imbalance's critical role in predictive analytics. Serving as a roadmap for practitioners and researchers dealing with imbalanced data, the findings guide model selection and data balancing strategies, enhancing classification performance despite class imbalance.
Citation: Lindani Dube, Tanja Verster. Enhancing classification performance in imbalanced datasets: A comparative analysis of machine learning models[J]. Data Science in Finance and Economics, 2023, 3(4): 354-379. doi: 10.3934/DSFE.2023021
[1] | Masoud Ghanbari, Jahan B Ghasemi, Amir M Mortazavian . Comparison of three sensory characterization methods based on consumer perception for the development of a novel functional cereal-based dessert. AIMS Agriculture and Food, 2017, 2(3): 258-278. doi: 10.3934/agrfood.2017.3.258 |
[2] | Kusumiyati Kusumiyati, Yuda Hadiwijaya, Wawan Sutari, Agus Arip Munawar . Global model for in-field monitoring of sugar content and color of melon pulp with comparative regression approach. AIMS Agriculture and Food, 2022, 7(2): 312-325. doi: 10.3934/agrfood.2022020 |
[3] | Marcelo Augusto de Carvalho, Cíntia Sorane Good Kitzberger, Altamara Viviane de Souza Sartori, Marta de Toledo Benassi, Maria Brígida dos Santos Scholz, Clandio Medeiros da Silva . Free choice profiling sensory analysis and principal component analysis as tools to support an apple breeding program. AIMS Agriculture and Food, 2020, 5(4): 769-784. doi: 10.3934/agrfood.2020.4.769 |
[4] | Nicole Roberta Giuggioli, Thais Mendes da Silva, Simone Dimitrova . Geographical indication (GI) branded quality: a study case on the homogeneity of the Carota Novella di Ispica Region. AIMS Agriculture and Food, 2021, 6(2): 538-550. doi: 10.3934/agrfood.2021031 |
[5] | Saima Latif, Muhammad Sohaib, Sanaullah Iqbal, Muhammad Hassan Mushtaq, Muhammad Tauseef Sultan . Comparative evaluation of nutritional composition, phytochemicals and sensorial attributes of lyophilized vs conventionally dried Grewia asiatica fruit pulp powder. AIMS Agriculture and Food, 2025, 10(1): 247-265. doi: 10.3934/agrfood.2025013 |
[6] | Nur Laylah, S. Salengke, Amran Laga, Supratomo Supratomo . Effects of the maturity level and pod conditioning period of cocoa pods on the changes of physicochemical properties of the beans of Sulawesi 2 (S2) cocoa clone. AIMS Agriculture and Food, 2023, 8(2): 615-636. doi: 10.3934/agrfood.2023034 |
[7] | Siti Hajar Othman, Nazirah Mohd Rosli, Norhazirah Nordin, Masturina Abdul Aziz . Formulation of crispy chicken burger patty batter: Properties and storage qualities. AIMS Agriculture and Food, 2022, 7(2): 426-443. doi: 10.3934/agrfood.2022027 |
[8] | Anna V. Babii, Anna L. Arkhipova, Irina N. Andreichenko, Artyom V. Brigida, Svetlana N. Kovalchuk . A TaqMan PCR assay for detection of DGAT1 K232A polymorphism in cattle. AIMS Agriculture and Food, 2018, 3(3): 306-312. doi: 10.3934/agrfood.2018.3.306 |
[9] | Stefano Puccio, Anna Perrone, Giuseppe Sortino, Giuseppe Gianguzzi, Carla Gentile, Vittorio Farina . Yield, pomological characteristics, bioactive compounds and antioxidant activity of Annona cherimola Mill. grown in mediterranean climate. AIMS Agriculture and Food, 2019, 4(3): 592-603. doi: 10.3934/agrfood.2019.3.592 |
[10] | Cíntia Sorane Good Kitzberger, David Pot, Pierre Marraccini, Luiz Filipe Protasio Pereira, Maria Brígida dos Santos Scholz . Flavor precursors and sensory attributes of coffee submitted to different post-harvest processing. AIMS Agriculture and Food, 2020, 5(4): 700-714. doi: 10.3934/agrfood.2020.4.700 |
In the realm of machine learning, where data-driven insights guide decision-making, addressing the challenges posed by class imbalance in datasets has emerged as a crucial concern. The effectiveness of classification algorithms hinges not only on their intrinsic capabilities but also on their adaptability to uneven class distributions, a common issue encountered across diverse domains. This study delves into the intricate interplay between varying class imbalance levels and the performance of ten distinct classification models, unravelling the critical impact of this imbalance on the landscape of predictive analytics. Results showed that random forest (RF) and decision tree (DT) models outperformed others, exhibiting robustness to class imbalance. Logistic regression (LR), stochastic gradient descent classifier (SGDC) and naïve Bayes (NB) models struggled with imbalanced datasets. Adaptive boosting (ADA), gradient boosting (GB), extreme gradient boosting (XGB), light gradient boosting machine (LGBM), and k-nearest neighbour (kNN) models improved with balanced data. Adaptive synthetic sampling (ADASYN) yielded more reliable predictions than the under-sampling (UNDER) technique. This study provides insights for practitioners and researchers dealing with imbalanced datasets, guiding model selection and data balancing techniques. RF and DT models demonstrate superior performance, while LR, SGDC and NB models have limitations. By leveraging the strengths of RF and DT models and addressing class imbalance, classification performance in imbalanced datasets can be enhanced. This study enriches credit risk modelling literature by revealing how class imbalance impacts default probability estimation. The research deepens our understanding of class imbalance's critical role in predictive analytics. Serving as a roadmap for practitioners and researchers dealing with imbalanced data, the findings guide model selection and data balancing strategies, enhancing classification performance despite class imbalance.
Among various fermented milk products, shrikhand occupies a unique place in the Indian sweet dishes. It is one of the major indigenous fermented milk products, prepared from lactic fermented curd and a semi-soft, healthful, delicious whole milk dessert popular in western part of India (Gujarat & Rajasthan). It is made with chakka (strained yoghurt/curd) which is finely mixed with sugar and flavoring agents [1,2], apart from fruits, nuts, sugar, cardamom, saffron and other spices. Although shrikhand is a concentrated source of milk fat, protein, calories, and minerals, the amount of sugar (more than 40%) and fat (8–12%) present in shrikhand limits its consumption. As a result, the calorie conscious consumers, diabetics and persons suffering with Coronary Heart Diseases (CHDs) cannot relish it [3]. Therefore, demand for reduced calorie and low-fat milk products are now becoming quite popular, as these can help the body to lower the risk of chronic degenerative or gastrointestinal issues [4]. In this context, stevia is a natural sweetener obtained from the leaf of Stevia rebaudiana, a perennial herb from the Asteraceae family plant. The species S. rebaudiana Bertoni, commonly known as sugar leaf, or simply stevia, is widely grown for its sweetness. Steviol glycosides (SGs) are the secondary metabolites responsible for the sweetness of stevia [5], which is much higher than sucrose. Apart from being a non-caloric sugar substitute, which is beneficial against diabetes disease, it improves digestion and prevents tooth enamel decay [6].
Recently, there has been an increased trend to fortify cultured milk products with fruit or vegetable pulps to enhance the versatility of flavor, texture, and color. The association of fruits and vegetables with cultured dairy products has endorsed healthy perception in the consumer mind, as these contain significant levels of biologically active components and important phytonutrients namely, vitamins, minerals, antioxidants, and dietary fibers that provide specific health benefits beyond the traditional nutrients [7]. Current evidence collectively demonstrates that fruit and vegetable intake reduce the risk of various types of cancers, CHD, hypertension, and possibly delayed onset of age-related indicators [8,9]. One such vegetable is carrot, which has intrinsic nutritional value, with rich dietary fiber, vitamins, and mineral contents, which make food products more functional and beneficial to health.
Although research on different quality parameters of shrikhand incorporating pulp from different fruits such as kiwi [10,11], apple [12], custard apple and stevia [13], papaya [14], mango [15], jamun [16], date [17], etc. have been conducted, no information is available for utilizing carrot pulp in shrikhand. Considering the demand for low fat and low sugar traditional dairy foods fortified with vegetable pulp, the present investigation was undertaken to standardize the milk fat and stevia level, optimize incorporation level of carrot pulp in chakka, and study the refrigerated storage quality of developed shrikhand at 3-day intervals up to 9 days.
Fresh and clean whole cow milk, obtained from Dairy Technology Section of ICAR-IVRI, Izatnagar (Uttar Pradesh), was analyzed for various parameters and for the preparation of shrikhand. Cow milk was passed through a cream separator to obtain skim milk and cream. The skim milk and cream were used to standardize the milk as per the experimental plan. Household curd was used as starter culture for the setting of Dahi. Good quality white crystalline food grade cane sugar was procured from the local supermarket. The sugar was ground in a laboratory grinder for mixing with chakka to manufacture shrikhand. Stevia powder (Zydus Wellness Products Ltd, Mumbai, India) and fresh carrots used for the study were purchased from the local supermarket. The carrots were washed and subjected to blanching in hot water maintaining a temperature of 60 ℃ for 15 min to deactivate the enzymes and prolong the storage period, and then were chopped and grinded in a domestic mixer for sufficient time, until a fine consistency of pulp was obtained.
Shrikhand was prepared by the traditional method suggested by Aneja et al. [18]. Whole cow milk was pre-heated to 40 ℃ before being subjected to the cream separator for the preparation of skimmed milk. Whole milk in different proportions (4.5, 3, 1.5, or 0.5%) was mixed to optimize fat percentage for the preparation of low fat shrikhand. The most acceptable product selected as per the sensory evaluation was used to carry out further studies.
Shrikhand was made following the previously mentioned procedure, with the addition of fine sugar at a 35% weight ratio to the chakka, serving as the control (C). The sugar content in shrikhand was subsequently substituted at three different levels with fine stevia powder, resulting in sugar-to-stevia ratios of 60:40, 40:60, and 20:80 for T1, T2, and T3, respectively. The selection of the optimal stevia level in chakka for the preparation of low-calorie shrikhand was based on sensory acceptability. Based on the sensory acceptability, shrinkand prepared with 60% sugar replacement with stevia was found comparatively better than T1 and T3.
Again, shrikhand was prepared using chakka, with the incorporation of carrot pulp at three distinct levels: 10, 20, and 30%, individually replacing chakka in the formulation. These variants were labeled as T1, T2, and T3, respectively. Similarly, shrikhand prepared from chakka without any carrot pulp was regarded as the control (C). The chakka was produced in larger quantities from cow's milk, with optimized levels of fat (1.5%) and stevia powder (60%). The extent of carrot pulp inclusion in the shrikhand was fine-tuned based on the results of sensory assessments, paving the way for further research.
The shrikhand containing the most preferred level of carrot pulp (20%) (T1) was compared with the control (C), which was prepared without any carrot pulp to replace the chakka. Both products were packaged in PET jars and stored at a refrigerated temperature of 4 ± 1 ℃. Over a period of 9 days, various parameters including physico-chemical factors such as pH, thiobarbituric acid reacting substances (TBARS) value, and water activity, sensory attributes, and microbiological quality (total plate count, coliform count, psychrophilic count, yeast and mold counts) were analyzed at regular intervals of three days (0, 3, 6, and 9 days) during the storage duration.
The pH of the milk and shrikhand samples was determined by using digital pH meter (pH tutor, Eutech instruments). The water activity of shrikhand samples were measured using a hand-held, portable digital water activity meter (Aqua lab dew point water activity meter 4TE, USA) as per the method adopted by Das et al. [19]. The titrable acidity of the milk and shrikhand samples was determined according to the method of AOAC [20]. The solid not fat (SNF) %, total solids (TS) % and specific gravity of milk was calculated as per the standard procedures of FSSAI [21] and AOAC [20], respectively.
The TBARS value of shrikhand was determined by using the distillation method [22] with suitable modifications. Ten grams of sample was mixed with 25 mL of pre-cooled 20% trichloroacetic acid (TCA) solution for 2 min. The content was then quantitatively transferred into a beaker by rinsing with 25 mL of chilled distilled water, well mixed and filtered through Whatman filter paper No. 1. Then, 3 mL of TCA extract (filtrate) was mixed with 3 mL of TBA reagent (0.005 M) in test tubes and placed in a water bath at 70˚C for 35 minutes. A blank sample was made by mixing 3 mL of 10% TCA and 3 mL of 0.005 M TBA reagent. Absorbance (O.D.) was measured at a fixed wavelength of 532 nm with a scanning range of 531–533 nm using a spectrophotometer (Spectramax, M5, USA). The TBARS value was calculated as mg malonaldehyde per kg of sample by multiplying O.D. value with a factor 5.2.
The moisture, protein, fat, and ash contents of the milk was estimated using a hot air oven, Kjeldahl assembly, Gerber centrifuge/Soxhlet extraction apparatus and Muffle furnace, respectively as per methods described by AOAC [20].
The total plate count (TPC), psychrophilic count, coliform count, and yeast and mold counts of the samples were enumerated following the methods as described by American Public Health Association [23].
An experienced sensory panel consisting of the scientists and post graduate students of the Division of Livestock Products Technology, ICAR-IVRI, Izatnagar (U.P.) evaluated the test samples for different sensory attributes, based on eight-point hedonic score card method [24]. The panelists were briefed about the nature of the experiments without disclosing the identity of the samples and were asked to rate them on an eight-point descriptive scale for different attributes like color and appearance, flavor, body and texture, sweetness, and over acceptability. Plain potable water was provided to rinse the mouth in between the samples.
The data collected from the experiments were consolidated and subjected to statistical analysis using SPSS-20 statistical software (Version 20, IBM, USA). Duplicate samples were taken for each parameter for which six samples were collected (except sensory evaluation). This entire experimental procedure was repeated three times, resulting in a total of six observations for all parameters. In the case of sensory evaluation, there were 18 observations. Duncan's multiple range tests were used to compare the means for significant differences. The statistical significance was determined at a 95% confidence level (p < 0.05).
The chemical analysis of the cow milk used in the present study is presented in Table 1. The data clearly suggests that the milk was of good quality and the percentage of total solid, fat, protein, and titrable acidity were within the limit of legal standards for cow milk as described by De [25].
Parameters | Percentage (%) |
Moisture | 87.83 ± 0.04 |
Fat | 4.57 ± 0.04 |
Protein | 3.25 ± 0.01 |
Solid not fat (SNF) | 7.57 ± 0.04 |
Total solid (TS) | 12.15 ± 0.03 |
Ash | 0.69 ± 0.01 |
Titrable acidity | 0.13 ± 0.001 |
The mean sensory scores of shrikhand prepared from milk with different fat levels viz. 4.5, 3.0, 1.5, and 0.5% are presented in Figure 1. The product prepared from milk with 0.5% milk (T3) fat showed significantly (p < 0.05) lower scores as compared to control (4.5%), T1 (3.0%), and T2 (1.5%). As expected, the control sample showed the highest mean score values for appearance and color, flavor, body and texture, sweetness, and overall acceptability. The sensory scores for the product with 1.5% fat showed no significant (p > 0.05) difference with product prepared from 3% fat milk and a direct relationship between fat content and overall acceptability of the product was observed. Similar to our study, Aykan et al. [26], Kim et al. [27], and Saleh et al. [28] have also prepared low-fat vanilla ice cream and yoghurt, respectively. In a study conducted by Reddam and Prabhakar [29], four different milk fat level (6.0, 4.5, 3.0, and 0.5%) were used for paneer preparation and according to hedonic scale and proximate analysis, paneer prepared from milk with initial 3.0% fat was acceptable with good hedonic points.
Various researchers have also prepared dairy products, including shrikhand using milk fat from buffalo, goat, cow, etc., and the use of milk fat up to 6% has been reported [13]. But apart from being expensive, the increased milk fat had no significant effect on acidity and pH of the final product. Therefore, considering the health benefits and sensory scores, the optimum milk fat level of 1.5% to prepare low fat shrikhand in this study seems to be appropriate.
To optimize the level of stevia powder to prepare low sugar shrikhand, different versions of shrikhand were prepared by replacing the sugar with stevia at three different levels viz. 40, 60, and 80% and subjected to sensory evaluation. Mean sensory scores of shrikhand with different levels of stevia powder are given in Figure 2. It was observed that the scores for important sensory attributes such as appearance and color, flavor, body and texture, sweetness, and overall acceptability were the highest for control (C) and decreased with increasing stevia levels. The product with 80% stevia (T3) showed significantly (p < 0.05) lower sensory scores. Significant differences (p < 0.05) in flavor and sweetness scores were observed in control and 40 % stevia (T1), while all other sensory parameters showed non-significant (p > 0.05) differences. Although, the scores were significantly (p < 0.05) lower for T2 and T3 with stevia levels 60% and 80% respectively as compared to control, the scores of T1 and T2 were comparable and above very good category. Hence, based on sensory scores and considering the effect of sugar on human health, the optimum level of stevia was adjudged as 60% to prepare the low fat, low sugar shrikhand.
Several researchers have used stevia as a sweetener to replace sugar in kesar peda [30], milk shake [31] and chocolate milk [32]. Replacing sugar in dietetic kulfi, Giri et al. [6] observed a decrease in the sensory perception with increasing the level of stevia. Replacing > 50% sugar with stevia resulted in bitterness, lack of brownish appearance and presence of icy texture in dietetic kulfi. Likewise, the sensory attributes were the best scored, when sucrose and stevia were added (1:1) in strawberry flavored yoghurt [33] suggesting that up to 50% sucrose replacements with stevia were more acceptable by the consumers.
Similarly, in an experiment conducted by Tondare and Hembade [15], sensory properties of amrakhand samples with sucrose and stevia leaf extract powder in the ratio of 30:70 was found to be better than control with 100:0 sucrose and stevia leaf powder. In an experiment conducted by Alizadeh [31], where five different treatments of fruit milk shakes were prepared with sucrose/stevia in the ratios of 100:0, 75:25, 50:50, 25:75, and 0:100, the recommended ratio of sucrose/stevia in beverage was 25:75. From these findings, it can be concluded that stevia is a good choice to develop low sucrose dairy products for health-conscious people. Our study suggests that 60% sugar replacement with stevia level had no adverse impact on sensorial acceptance of the product.
This study was conducted to optimize the level of incorporation of carrot pulp for the preparation of shrikhand by replacing chakka at three different levels (10, 20, and 30%). Mean sensory scores of shrikhand with different levels of carrot pulp are presented in Figure 3. No significant difference (p > 0.05) was observed among the appearance and color scores of the control (C) and product with 20% carrot pulp (T2). Although no significant (p > 0.05) difference was observed among the treatment product with 30% carrot pulp (T3) and product with 10% carrot pulp (T2), but T3 in comparison showed the lowest mean appearance and color score.
The mean scores for the flavor of T2 were significantly (p < 0.05) higher compared to the control and other treatments, whereas. the flavor score of control t and product with 10% carrot pulp (T1) had no significant difference (p > 0.05). The treatment product with 30% carrot pulp (T3) showed significantly (p < 0.05) lower mean scores for flavor and this may be attributed to high level of carrot pulp that masked the original flavor of the product. Likewise, the mean scores for body and texture of shrikhand with 10% carrot pulp (T1) was comparable to that of the control (C) and 20% carrot pulp (T2). On the other hand, the product with 30% carrot pulp (T3) showed significantly lower (p < 0.05) mean scores for the body and texture. In a study, Kumar et al. [34] have also reported a decrease in texture of shrikhand with increased levels of apple pulp.
The mean scores for sweetness of shrikhand treatment product with 10% and 20% carrot pulp were comparable to that of the control product, whereas the treatment product with 30% carrot pulp (T3) showed significantly lower (p < 0.05) scores for sweetness. The results of this study are in agreement with the findings of El-Said et al. [35], as the workers also reported a decreased sweetness score in stirred yoghurt, when pomegranate peel extracts were added at the highest level (35%).
Mean scores for the overall acceptability of product showed no significant difference (p < 0.05) between control and treatment product with 20% carrot pulp, but there was significant difference (p < 0.05) among the treatment products. Furthermore, the mean scores for the treatment product with 30% carrot pulp showed significantly lower (p < 0.05) overall acceptability scores. Hence, based on sensory scores, the optimum level of replacement of chakka with carrot pulp was judged to be 20%.
Several reports are available on blending of fruit or vegetable pulps in the preparation of dairy products. Excellent quality carrot-yoghurt could be prepared by blending milk in different proportions with 5–20% carrot juice before fermentation [36]. Nigam et al. [37] conducted a study to prepare shrikhand by incorporating papaya pulp at 20, 40, and 60% replacing chakka to increase the nutritional quality and overall acceptability. According to the researchers, shrikhand prepared with 20% level of papaya pulp was the most acceptable. In a similar line of study, Suryawanshi et al. [14] blended 10, 20, and 30% papaya pulp in probiotic shrikhand and the most acceptable sensory scores were obtained with 20% papaya pulp incorporated shrikhand. Studying the development of shrikhand incorporated with kiwi (Actinidia deliciosa) pulp, Kedaree et al. [38] concluded that shrikhand blended with 15% kiwi pulp had better overall acceptability than the other treatments. Development of functional shrikhand incorporating aqueous extracts (20%) of orange fruit peels had a better sensory evaluation report compared to the control [39]. The overall acceptability of shrikhand blended with 20% unripe banana pulp was better on the basis of sensory score as compared to other treatments and control [40]. Assessing the incorporation of various levels of custard apple pulp, Kamble [13] concluded that shrikhand prepared with 10% custard apple pulp had improved sensory attributes compared to others.
Development of goat milk shrikhand incorporated with 25 % apple fruit pulp improved the texture and overall acceptability scores compared to other treatments [12]. However, Joshna et al. [41] recorded increased scores of all sensory attributes and the highest acceptability of fruit based chakka desserts incorporated with 70 % sweetened fruit pulp of fig, star fruit, or blueberry over 30 or 50%. From this, it can be deduced that the level of incorporation of pulp in different desserts depends on the intensity of fruit or vegetable flavor and overall sensory acceptability of the end product.
The storage quality of shrikhand with optimum level of carrot pulp (T-20%) was evaluated in terms of physico-chemical, microbiological and sensory characteristics, and compared with control (C). Both the products (C and T) were packed in PET jars and analyzed every three days, for nine days during storage at refrigerated temperature (4 ± 1 ℃).
The changes in physico-chemical characteristics of the control and product with optimum level of carrot pulp (20%) were evaluated over the period of storage and are presented in Table 2.
Refrigerated storage period (Days) | ||||
pH | ||||
Treatment | 0day | 3rd day | 6th day | 9th day |
C | 4.23 ± 0.005aB | 4.16 ± 0.006Bb | 4.11 ± 0.006cB | 4.07 ± 0.003dB |
T | 4.28 ± 0.003aA | 4.19 ± 0.007Ba | 4.15 ± 0.008cA | 4.10 ± 0.004dA |
Water activity (aw) | ||||
C | 0.963 ± 0.007dB | 0.971 ± 0.007cB | 0.977 ± 0.006bB | 0.982 ± 0.007aB |
T | 0.973 ± 0.004dA | 0.982 ± 0.007cA | 0.987 ± 0.003bA | 0.991 ± 0.004aA |
TBARS value (mg malonaldehyde/kg) | ||||
C | 0.387 ± 0.004dA | 0.498 ± 0.002cA | 0.535 ± 0.001bA | 0.582 ± 0.001aA |
T | 0.384 ± 0.002dB | 0.424 ± 0.003Cb | 0.445 ± 0.002bB | 0.497 ± 0.001aB |
*Mean with different superscripts row wise (small letters) and column wise (capital letters) differ significantly (p < 0.05); n = 6; Control: 100% chakka; T: with 20% carrot pulp |
pH
The average pH of control and treatment product indicated a decreasing trend from 0 day to 9th day of storage period (Table 2). However, the pH value of shrikhand with carrot pulp remained significantly (p < 0.05) higher than the control during the entire period of storage. Para et al. [42] also reported a decrease in the pH of shrikhand during storage. The lowering of pH might have been due to lipolysis that leads to fatty acid release during storage. Furthermore, the acidic nature of the ingredients used might have contributed to the increase in pH level.
Water activity
The mean value for water activity (aw) in both the products showed a gradual increasing trend with a significant increase (p < 0.05) from 3rd day to 9th day of storage (Table 2). The aw of the control product remained significantly lower (p < 0.05) than treatment product throughout the storage period. This may be due to the high moisture content of carrot pulp in shrikhand in comparison to chakka that increased the moisture content of treatment product.
TBARS value
The TBARS values were evaluated to assess lipid oxidation in control (C) and treatment product with 20 % carrot pulp (T). The mean TBARS value of control and treatment products indicated an increasing trend. The increase in TBARS value on storage might be attributed to increased lipid oxidation and production of volatile metabolites during storage. However, the TBARS value of treated shrikhand was significantly (p < 0.05) lower than control samples during storage study, which may be due to antioxidant effect of carrot pulp. In a study by Kumar et al. [34], an increase in TBARS value of shrikhand was also observed, but the increase was lower in treatment product incorporated with apple pulp, which was attributed to antioxidant effect of apple pulp.
The microbiological quality (total plate count, psychrophilic count, coliform count, yeast and mold count) for control and carrot pulp incorporated shrikhand was analyzed on 0, 3rd, 6th, and 9th day of storage at refrigerated temperature (4 ± 1 ℃) and the mean ± SE values are presented in Table 3.
Refrigerated storage period (Days) | ||||
Total plate count (X 104 cfu/g) | ||||
Treatment | 0 day | 3rd day | 6th day | 9th day |
C | 2.06 ± 0.16d | 2.42 ± 0.07c | 2.85 ± 0.04bA | 3.16 ± 0.02aA |
T | 2.05 ± 0.15d | 2.40 ± 0.05c | 2.81 ± 0.07bB | 3.08 ± 0.03aB |
Psychrophilic count (log10 cfu/g) | ||||
C | ND | ND | ND | 1.96 ± 0.01b |
T | ND | ND | ND | 1.92 ± 0.02b |
Coliform count (log10 cfu/g) | ||||
C | ND | ND | ND | 1.008 ± 0.02a |
T | ND | ND | ND | 1.005 ± 0.02a |
Yeast & mold count (log10 cfu/g) | ||||
C | ND | ND | ND | 0.793 ± 0.03a |
T | ND | ND | ND | 0.791 ± 0.03a |
*Mean with different superscripts row wise (small letters) and column wise (capital letters) differ significantly (p < 0.05); n = 6; Control: 100% chakka; T: with 20% carrot pulp. |
Total plate count
The total plate count followed a gradual increasing trend in the treatment product as well as the control, with an increase in storage period. The TPC of the control and treatment product were comparable until the 3rd day of storage, but control showed significantly (p < 0.05) higher TPC compared to the treatment product from 6th day onwards.
Similar increase in mean values of TPC with increase in days of storage under refrigeration have been reported by different workers in shrikhand prepared by incorporating apple pulp [34] and kiwi pulp [11] and also dietetic Kashmiri saffron phirne prepared from reconstituted skim milk [43].
Psychrophilic count
The psychrophiles were not detected up to the 6th day of storage in the control and the treatment product. It may be attributed to retardation of the log phase, because of reduced metabolic rate due to sudden change in the physical environment. On the 9th day of storage, although psychrophiles were detected, there was no significant difference (p < 0.05) in counts between the control and treatment products. Absence of psychrophiles in fresh samples and its consistent increase with increase in days of storage aligns with the reports in various dairy products such as paneer [19], milk nuggets [44], dietetic Kashmiri saffron phirne prepared from reconstituted skim milk [43], and shrikhand prepared by incorporating apple pulp [34].
Coliform count
Coliforms were not detected up to the 6th day of storage in the control and treatment product. However, they were detected on the 9th day of storage. It might be due to thermal injury to the bacterial cells during the boiling of the milk before preparation of the product and the fact that some injured bacterial cells might have rejuvenated after lag phase and then entered active multiplication phase. A similar finding was reported by Kumar et al. [34] that coliform was not detected in apple pulp incorporated shrikhand during the entire period of storage. Bhat et al. [43] also reported zero coliform in Kashmiri saffron phirne during refrigerated storage. Das et al. [19] and Buch et al. [45] also observed no coliform in paneer treated with different natural preservatives during refrigerated storage.
Yeast and mold count
Yeast and mold were not detected up to the 6th day of storage in the control and treatment product. Although detected on the 9th day of storage, no significant difference (p < 0.05) was found between the yeast and mold counts of the control and treatment product. Thakur et al. [46] reported yeast and mold count in fruit fortified shrikhand with different levels of mango pulp in the range of 0.93–1.3l cfu/g. A similar finding was also reported by Kahate et al.[40] that there was a significantly (p < 0.05) increasing yeast and mold count observed in unripe banana incorporated shrikhand during storage period.
The mean sensory scores of shrikhand incorporated with 20% carrot pulp (T) and the control (C) during storage are presented in Table 4. The mean score for appearance and color showed a significantly (p < 0.05) decreasing trend with increasing storage days for both the control and product with 20% carrot pulp. However, the score of the treatment product was significantly (p < 0.05) higher than the control product on all the days of storage. Likewise, the mean score of flavor for the control and treatment product, although decreasing over the storage period, was significantly (p < 0.05) higher for the treatment product than control. This may be due to characteristic flavor imparted by carrot pulp incorporation in shrikhand.
The mean score for body and texture of the control and treatment product showed a gradual and significant (p < 0.05) decreasing trend with increase in storage days, but the control product had higher body and texture scores. Although non-significant (p > 0.05), the decrease in the texture score of treated products could be due to an increase in moisture content of shrikhand with optimum level of carrot pulp.
The mean score of sweetness for control and treatment product on 0 day of storage had no significant difference (p > 0.05). However, the control and treatment product showed highly significant difference (p < 0.05) from the 3rd day onwards. The mean scores for sweetness of the control and treatment product showed significantly (p < 0.05) decreasing trends with increase in storage days.
Mean scores for overall acceptability of the control and treatment product showed a gradual and significant (p < 0.05) decreasing trend throughout the storage days. The mean scores for overall acceptability of the control and treatment product showed no significant (p > 0.05) difference up to the 3rd day of storage, and after that it was significantly (p > 0.05) higher for the treatment product.
Das et al. [19], Nigam et al. [37], and Bhat et al. [43] also reported similar declines in the sensory parameters of various dairy products during refrigerated storage. Patel et al. [47] reported that overall acceptability of chakka decreased with an increase in storage period due to deterioration of flavor. Likewise, Joshna et al. [41] recorded sensory attributes of fruit based chakka desserts within the acceptable range during storage study for about 10 days at 5 ℃. From our study, it can be concluded that shrikhand prepared with 1.5% fat milk and stevia as 60% replacement of sugar and incorporated with 20% carrot pulp may be stored up to 9 days under refrigeration without marked changes in the quality.
Refrigerated storage period (Days) | ||||
Appearance and color | ||||
Treatment | 0 day | 3rd day | 6th day | 9th day |
C | 7.13 ± 0.03Ab | 6.93 ± 0.04bB | 6.54 ± 0.03cB | 6.05 ± 0.02dB |
T | 7.21 ± 0.03aA | 7.01 ± 0.01bA | 6.68 ± 0.02A | 6.34 ± 0.02dA |
Flavor | ||||
C | 7.02 ± 0.02aB | 6.86 ± 0.05bB | 6.36 ± 0.04cB | 6.02 ± 0.01dB |
T | 7.10 ± 0.02aA | 6.94 ± 0.04bA | 6.61 ± 0.03cA | 6.20 ± 0.03dA |
Body and texture | ||||
C | 7.11 ± 0.03aA | 6.91 ± 0.04bA | 6.66 ± 0.03cA | 6.45 ± 0.05dA |
T | 7.04 ± 0.05aB | 6.88 ± 0.07bB | 6.54 ± 0.08cB | 6.30 ± 0.09dB |
Sweetness | ||||
C | 7.05 ± 0.02aA | 6.94 ± 0.02bB | 6.79 ± 0.04cB | 6.38 ± 0.03dB |
T | 7.09 ± 0.03aA | 7.01 ± 0.03aA | 6.84 ± 0.03bA | 6.58 ± 0.02cA |
Overall acceptability | ||||
C | 6.97 ± 0.04aA | 6.87 ± 0.04bA | 6.57 ± 0.03cB | 6.19 ± 0.03bB |
T | 7.1 ± 0.04aA | 6.91 ± 0.03bA | 6.65 ± 0.02cA | 6.23 ± 0.03dA |
*Mean with different superscripts row wise (small letters) and column wise (capital letters) differ significantly (p < 0.05); n = 6; Control: 100% chakka; T1: with 20% carrot pulp. |
Considering the demand for low fat and low sugar traditional dairy foods, the present investigation was undertaken to standardize the technology for the production of low fat and low sugar shrikhand. Based on sensory scores, 1.5% fat milk was judged to be optimum. Furthermore, sugar in shrikhand with an optimum level of milk fat was replaced with different levels of stevia powder, and 60% replacement of sugar with stevia was judged to be the optimum. Shrikhand prepared with 1.5% fat milk and 60% sugar replacement with stevia was incorporated with different levels of carrot pulp separately, replacing the chakka, and 20% incorporation level was found to be optimum for each separately. The control and the treatment products with 20% carrot pulp were evaluated for various physico-chemical, sensory, and microbial characteristics. Sensory evaluation revealed that mean scores for color, appearance and overall acceptability were significantly higher in carrot pulp blended shrikhand. Furthermore, microbiological quality parameters were within permissible limits for both the products. From this study, it can be concluded that the developed shrikhand can be stored at refrigerated storage temperature for 9 days without marked change in the quality, but shrikhand with 20% carrot pulp may be preferred because of the value addition it offers, enhancing functional properties and acceptability of the product.
The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.
The authors are thankful to the Director, ICAR- Indian Veterinary Research Institute, Izatnagar and Station In-charge, Eastern Regional Station, ICAR-Indian Veterinary Research Institute, Kolkata, India for providing necessary facilities in conducting this study.
All authors declare no conflicts of interest in this paper.
[1] |
Alija S, Beqiri E, Gaafar AS, et al. (2023) Predicting students performance using supervised machine learning based on imbalanced dataset and wrapper feature selection. Informatica 47. https://doi.org/10.31449/inf.v47i1.4519 doi: 10.31449/inf.v47i1.4519
![]() |
[2] |
Aljedaani W, Rustam F, Mkaouer MW, et al. (2022) Sentiment analysis on twitter data integrating textblob and deep learning models: The case of us airline industry. Knowl-Based Syst 255: 109780. https://doi.org/10.1016/j.knosys.2022.109780 doi: 10.1016/j.knosys.2022.109780
![]() |
[3] | Anguita D, Ghelardoni L, Ghio A, et al. (2012) The'k'in k-fold cross validation. in 'ESANN', 441–446. |
[4] |
Bentéjac C, Csörgő A, Martínez-Muñoz G (2021) A comparative analysis of gradient boosting algorithms. Artif Intell Rev 54: 1937–1967. https://doi.org/10.1007/s10462-020-09896-5 doi: 10.1007/s10462-020-09896-5
![]() |
[5] |
Booth A, Gerding E, McGroarty F (2015) Performance-weighted ensembles of random forests for predicting price impact. Quant Financ 15: 1823–1835. https://doi.org/10.1080/14697688.2014.983539 doi: 10.1080/14697688.2014.983539
![]() |
[6] | Breeden J (2021) A survey of machine learning in credit risk. J Credit Risk 17. https://ssrn.com/abstract = 3946261 |
[7] |
Breiman L (2001) Random forests. Mach learn 45: 5–32. https://doi.org/10.1023/A:1010933404324 doi: 10.1023/A:1010933404324
![]() |
[8] | Breiman L, Friedman J, Olshen R, et al. (1984) Classification and regression trees (wadsworth, belmont, ca). 13: 978–0412048418. |
[9] |
Calderoni L, Ferrara M, Franco A, et al. (2015) Indoor localization in a hospital environment using random forest classifiers. Expert Syst Appl 42: 125–134. https://doi.org/10.1016/j.eswa.2014.07.042 doi: 10.1016/j.eswa.2014.07.042
![]() |
[10] | Cawley GC, Talbot NL (2010) On over-fitting in model selection and subsequent selection bias in performance evaluation. J Mach Learn Res 11: 2079–2107. |
[11] | Chen T, Guestrin C (2016) Xgboost: A scalable tree boosting system, in Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, 785–794. |
[12] |
Chicco D, Jurman G (2020) The advantages of the matthews correlation coefficient (mcc) over f1 score and accuracy in binary classification evaluation. BMC Genomics 21: 1–13. https://doi.org/10.1186/s12864-019-6413-7 doi: 10.1186/s12864-019-6413-7
![]() |
[13] |
Das S, Datta S, Chaudhuri BB (2018) Handling data irregularities in classification: Foundations, trends, and future challenges. Pattern Recogn 81: 674–693. https://doi.org/10.1016/j.patcog.2018.03.008 doi: 10.1016/j.patcog.2018.03.008
![]() |
[14] | De Campos LM, Cano A, Castellano JG, et al. (2011) Bayesian networks classifiers for gene-expression data, in 2011 11th International Conference on Intelligent Systems Design and Applications, IEEE, 1200–1206. https://doi.org/10.1109/ISDA.2011.6121822 |
[15] |
Deng M, Chen J, Huang J, et al. (2018) Agricultural drought risk evaluation based on an optimized comprehensive index system. Sustainability 10: 3465. https://doi.org/10.3390/su10103465 doi: 10.3390/su10103465
![]() |
[16] | Dhieb N, Ghazzai H, Besbes H, et al. (2019) Extreme gradient boosting machine learning algorithm for safe auto insurance operations, in 2019 IEEE international conference on vehicular electronics and safety (ICVES), IEEE, 1–5. https://doi.org/10.1109/ICVES.2019.8906396 |
[17] |
Dorogush AV, Ershov V, Gulin A (2018) Catboost: gradient boosting with categorical features support. arXiv preprint. https://doi.org/10.48550/arXiv.1810.11363 doi: 10.48550/arXiv.1810.11363
![]() |
[18] | Fayyad UM, Irani KB (1992) The attribute selection problem in decision tree generation, in 'AAAI', 104–110. |
[19] | Fernando KRM, Tsokos CP (2021) Dynamically weighted balanced loss: class imbalanced learning and confidence calibration of deep neural networks. IEEE T Neur Net Learn Syst 33: 2940–2951. |
[20] | Granström D, Abrahamsson J (2019) Loan default prediction using supervised machine learning algorithms. |
[21] | Han J, Kamber M, Pei J (2012) Data mining concepts and techniques third edition, University of Illinois at Urbana-Champaign Micheline Kamber Jian Pei Simon Fraser University. |
[22] | Ho TK (1995) Random decision forests, in Proceedings of 3rd international conference on document analysis and recognition, IEEE, 1: 278–282. |
[23] | Kaggle (2023) Give me some credit. Available from: https://www.kaggle.com/competitions/GiveMeSomeCredit/dataselect = cs-training.csv. Accessed: 2023-02-05. |
[24] | Ke G, Meng Q, Finley T, et al. (2017) Lightgbm: A highly efficient gradient boosting decision tree. Adv Neural Inf Process Syst 30. |
[25] | Kelleher JD, Mac Namee B, D'arcy A (2020) Fundamentals of machine learning for predictive data analytics: algorithms, worked examples, and case studies, MIT press. |
[26] |
Khemakhem S, Boujelbene Y (2018) Predicting credit risk on the basis of financial and non-financial variables and data mining. Rev Account Financ 17: 316–340. https://doi.org/10.1108/RAF-07-2017-0143 doi: 10.1108/RAF-07-2017-0143
![]() |
[27] |
Laradji IH, Alshayeb M, Ghouti L (2015) Software defect prediction using ensemble learning on selected features. Inform Software Tech 58: 388–402. https://doi.org/10.1016/j.infsof.2014.07.005 doi: 10.1016/j.infsof.2014.07.005
![]() |
[28] |
Leo M, Sharma S, Maddulety K (2019) Machine learning in banking risk management: A literature review. Risks 7: 29. https://doi.org/10.3390/risks7010029 doi: 10.3390/risks7010029
![]() |
[29] |
Li K, Xu H, Liu X (2022) Analysis and visualization of accidents severity based on lightgbm-tpe. Chaos, Solitons Fract 157: 111987. https://doi.org/10.1016/j.chaos.2022.111987 doi: 10.1016/j.chaos.2022.111987
![]() |
[30] |
Liu L, Li P, Chu M, et al. (2021) Stochastic gradient support vector machine with local structural information for pattern recognition. Int J Mach Learn Cybe 12: 2237–2254. https://doi.org/10.1007/s13042-021-01303-x doi: 10.1007/s13042-021-01303-x
![]() |
[31] | Liu W, Chawla S, Cieslak DA, et al. (2010) A robust decision tree algorithm for imbalanced data sets, inProceedings of the 2010 SIAM International Conference on Data Mining, SIAM, 766–777. |
[32] | Lokeswari N, Amaravathi K (2018) Comparative study of classification algorithms in sentiment analysis. Int Res J Sci Eng Technol 4: 31–39. |
[33] | Mitchell TM, Mitchell TM (1997) Machine learning, 1: McGraw-hill New York. |
[34] |
Ogunleye A, Wang QG (2019) Xgboost model for chronic kidney disease diagnosis. IEEE/ACM T Comput Bi 17: 2131–2140. https://doi.org/10.1109/TCBB.2019.2911071 doi: 10.1109/TCBB.2019.2911071
![]() |
[35] |
Okey OD, Maidin SS, Adasme P, et al. (2022) Boostedenml: Efficient technique for detecting cyberattacks in iot systems using boosted ensemble machine learning. Sensors 22: 7409. https://doi.org/10.3390/s22197409 doi: 10.3390/s22197409
![]() |
[36] | Padmaja TM, Dhulipalla N, Bapi RS, et al. (2007) Unbalanced data classification using extreme outlier elimination and sampling techniques for fraud detection. in 15th International Conference on Advanced Computing and Communications (ADCOM 2007), IEEE, 511–516. https://doi.org/10.1109/ADCOM.2007.74 |
[37] | Patro S, Sahu KK (2015) Normalization: A preprocessing stage. arXiv preprint arXiv: 1503.06462. https://doi.org/10.48550/arXiv.1503.06462 |
[38] | Rubin DB (1976) Inference and missing data. Biometrika 63: 581–592. |
[39] |
Schafer JL, Graham JW (2002) Missing data: our view of the state of the art. Psychol Methods 7: 147. https://doi.org/10.1037/1082-989X.7.2.147 doi: 10.1037/1082-989X.7.2.147
![]() |
[40] | Singhal Y, Jain A, Batra S, et al. (2018) Review of bagging and boosting classification performance on unbalanced binary classification, in 2018 IEEE 8th International Advance Computing Conference (IACC), IEEE, 338–343. https://doi.org/10.1109/IADCC.2018.8692138 |
[41] |
Stephens D, Diesing M (2014) A comparison of supervised classification methods for the prediction of substrate type using multibeam acoustic and legacy grain-size data. PloS One 9: e93950. https://doi.org/10.1371/journal.pone.0093950 doi: 10.1371/journal.pone.0093950
![]() |
[42] |
Sun J, Lang J, Fujita H, et al. (2018) Imbalanced enterprise credit evaluation with dte-sbd: Decision tree ensemble based on smote and bagging with differentiated sampling rates. Inform Sci 425: 76–91. https://doi.org/10.1016/j.ins.2017.10.017 doi: 10.1016/j.ins.2017.10.017
![]() |
[43] |
Thabtah F, Hammoud S, Kamalov F, et al. (2020) Data imbalance in classification: Experimental evaluation. Inform Sci 513: 429–441. https://doi.org/10.1016/j.ins.2019.11.004 doi: 10.1016/j.ins.2019.11.004
![]() |
[44] | Wilson DR, Martinez TR (1997) Improved heterogeneous distance functions. J Artif Intell Res 6: 1–34. |
[45] | Yao Z, Ruzzo WL (2006) A regression-based k nearest neighbor algorithm for gene function prediction from heterogeneous data, BMC Bioinformatics, BioMed Central, 7: 1–11. https://doi.org/10.1186/1471-2105-7-S1-S11 |
[46] |
Zhang C, Liu C, Zhang X, et al. (2017) An up-to-date comparison of state-of-the-art classification algorithms. Expert Syst Appl 82: 128–150. https://doi.org/10.1016/j.eswa.2017.04.003 doi: 10.1016/j.eswa.2017.04.003
![]() |
[47] |
Zhou L, Wang H (2012) Loan default prediction on large imbalanced data using random forests. TELKOMNIKA Indonesian J Electr Eng 10: 1519–1525. https://doi.org/10.11591/telkomnika.v10i6.1323 doi: 10.11591/telkomnika.v10i6.1323
![]() |
1. | Manvik Joshi, Kamalesh Kumar Meena, Arun Kumar, Sunil Meena, Optimization of a novel probiotic-fermented pearl millet-based strained yoghurt-like functional dessert: physicochemical, microbial and sensory characterization, 2025, 2753-8095, 10.1039/D5FB00001G |
Parameters | Percentage (%) |
Moisture | 87.83 ± 0.04 |
Fat | 4.57 ± 0.04 |
Protein | 3.25 ± 0.01 |
Solid not fat (SNF) | 7.57 ± 0.04 |
Total solid (TS) | 12.15 ± 0.03 |
Ash | 0.69 ± 0.01 |
Titrable acidity | 0.13 ± 0.001 |
Refrigerated storage period (Days) | ||||
pH | ||||
Treatment | 0day | 3rd day | 6th day | 9th day |
C | 4.23 ± 0.005aB | 4.16 ± 0.006Bb | 4.11 ± 0.006cB | 4.07 ± 0.003dB |
T | 4.28 ± 0.003aA | 4.19 ± 0.007Ba | 4.15 ± 0.008cA | 4.10 ± 0.004dA |
Water activity (aw) | ||||
C | 0.963 ± 0.007dB | 0.971 ± 0.007cB | 0.977 ± 0.006bB | 0.982 ± 0.007aB |
T | 0.973 ± 0.004dA | 0.982 ± 0.007cA | 0.987 ± 0.003bA | 0.991 ± 0.004aA |
TBARS value (mg malonaldehyde/kg) | ||||
C | 0.387 ± 0.004dA | 0.498 ± 0.002cA | 0.535 ± 0.001bA | 0.582 ± 0.001aA |
T | 0.384 ± 0.002dB | 0.424 ± 0.003Cb | 0.445 ± 0.002bB | 0.497 ± 0.001aB |
*Mean with different superscripts row wise (small letters) and column wise (capital letters) differ significantly (p < 0.05); n = 6; Control: 100% chakka; T: with 20% carrot pulp |
Refrigerated storage period (Days) | ||||
Total plate count (X 104 cfu/g) | ||||
Treatment | 0 day | 3rd day | 6th day | 9th day |
C | 2.06 ± 0.16d | 2.42 ± 0.07c | 2.85 ± 0.04bA | 3.16 ± 0.02aA |
T | 2.05 ± 0.15d | 2.40 ± 0.05c | 2.81 ± 0.07bB | 3.08 ± 0.03aB |
Psychrophilic count (log10 cfu/g) | ||||
C | ND | ND | ND | 1.96 ± 0.01b |
T | ND | ND | ND | 1.92 ± 0.02b |
Coliform count (log10 cfu/g) | ||||
C | ND | ND | ND | 1.008 ± 0.02a |
T | ND | ND | ND | 1.005 ± 0.02a |
Yeast & mold count (log10 cfu/g) | ||||
C | ND | ND | ND | 0.793 ± 0.03a |
T | ND | ND | ND | 0.791 ± 0.03a |
*Mean with different superscripts row wise (small letters) and column wise (capital letters) differ significantly (p < 0.05); n = 6; Control: 100% chakka; T: with 20% carrot pulp. |
Refrigerated storage period (Days) | ||||
Appearance and color | ||||
Treatment | 0 day | 3rd day | 6th day | 9th day |
C | 7.13 ± 0.03Ab | 6.93 ± 0.04bB | 6.54 ± 0.03cB | 6.05 ± 0.02dB |
T | 7.21 ± 0.03aA | 7.01 ± 0.01bA | 6.68 ± 0.02A | 6.34 ± 0.02dA |
Flavor | ||||
C | 7.02 ± 0.02aB | 6.86 ± 0.05bB | 6.36 ± 0.04cB | 6.02 ± 0.01dB |
T | 7.10 ± 0.02aA | 6.94 ± 0.04bA | 6.61 ± 0.03cA | 6.20 ± 0.03dA |
Body and texture | ||||
C | 7.11 ± 0.03aA | 6.91 ± 0.04bA | 6.66 ± 0.03cA | 6.45 ± 0.05dA |
T | 7.04 ± 0.05aB | 6.88 ± 0.07bB | 6.54 ± 0.08cB | 6.30 ± 0.09dB |
Sweetness | ||||
C | 7.05 ± 0.02aA | 6.94 ± 0.02bB | 6.79 ± 0.04cB | 6.38 ± 0.03dB |
T | 7.09 ± 0.03aA | 7.01 ± 0.03aA | 6.84 ± 0.03bA | 6.58 ± 0.02cA |
Overall acceptability | ||||
C | 6.97 ± 0.04aA | 6.87 ± 0.04bA | 6.57 ± 0.03cB | 6.19 ± 0.03bB |
T | 7.1 ± 0.04aA | 6.91 ± 0.03bA | 6.65 ± 0.02cA | 6.23 ± 0.03dA |
*Mean with different superscripts row wise (small letters) and column wise (capital letters) differ significantly (p < 0.05); n = 6; Control: 100% chakka; T1: with 20% carrot pulp. |
Parameters | Percentage (%) |
Moisture | 87.83 ± 0.04 |
Fat | 4.57 ± 0.04 |
Protein | 3.25 ± 0.01 |
Solid not fat (SNF) | 7.57 ± 0.04 |
Total solid (TS) | 12.15 ± 0.03 |
Ash | 0.69 ± 0.01 |
Titrable acidity | 0.13 ± 0.001 |
Refrigerated storage period (Days) | ||||
pH | ||||
Treatment | 0day | 3rd day | 6th day | 9th day |
C | 4.23 ± 0.005aB | 4.16 ± 0.006Bb | 4.11 ± 0.006cB | 4.07 ± 0.003dB |
T | 4.28 ± 0.003aA | 4.19 ± 0.007Ba | 4.15 ± 0.008cA | 4.10 ± 0.004dA |
Water activity (aw) | ||||
C | 0.963 ± 0.007dB | 0.971 ± 0.007cB | 0.977 ± 0.006bB | 0.982 ± 0.007aB |
T | 0.973 ± 0.004dA | 0.982 ± 0.007cA | 0.987 ± 0.003bA | 0.991 ± 0.004aA |
TBARS value (mg malonaldehyde/kg) | ||||
C | 0.387 ± 0.004dA | 0.498 ± 0.002cA | 0.535 ± 0.001bA | 0.582 ± 0.001aA |
T | 0.384 ± 0.002dB | 0.424 ± 0.003Cb | 0.445 ± 0.002bB | 0.497 ± 0.001aB |
*Mean with different superscripts row wise (small letters) and column wise (capital letters) differ significantly (p < 0.05); n = 6; Control: 100% chakka; T: with 20% carrot pulp |
Refrigerated storage period (Days) | ||||
Total plate count (X 104 cfu/g) | ||||
Treatment | 0 day | 3rd day | 6th day | 9th day |
C | 2.06 ± 0.16d | 2.42 ± 0.07c | 2.85 ± 0.04bA | 3.16 ± 0.02aA |
T | 2.05 ± 0.15d | 2.40 ± 0.05c | 2.81 ± 0.07bB | 3.08 ± 0.03aB |
Psychrophilic count (log10 cfu/g) | ||||
C | ND | ND | ND | 1.96 ± 0.01b |
T | ND | ND | ND | 1.92 ± 0.02b |
Coliform count (log10 cfu/g) | ||||
C | ND | ND | ND | 1.008 ± 0.02a |
T | ND | ND | ND | 1.005 ± 0.02a |
Yeast & mold count (log10 cfu/g) | ||||
C | ND | ND | ND | 0.793 ± 0.03a |
T | ND | ND | ND | 0.791 ± 0.03a |
*Mean with different superscripts row wise (small letters) and column wise (capital letters) differ significantly (p < 0.05); n = 6; Control: 100% chakka; T: with 20% carrot pulp. |
Refrigerated storage period (Days) | ||||
Appearance and color | ||||
Treatment | 0 day | 3rd day | 6th day | 9th day |
C | 7.13 ± 0.03Ab | 6.93 ± 0.04bB | 6.54 ± 0.03cB | 6.05 ± 0.02dB |
T | 7.21 ± 0.03aA | 7.01 ± 0.01bA | 6.68 ± 0.02A | 6.34 ± 0.02dA |
Flavor | ||||
C | 7.02 ± 0.02aB | 6.86 ± 0.05bB | 6.36 ± 0.04cB | 6.02 ± 0.01dB |
T | 7.10 ± 0.02aA | 6.94 ± 0.04bA | 6.61 ± 0.03cA | 6.20 ± 0.03dA |
Body and texture | ||||
C | 7.11 ± 0.03aA | 6.91 ± 0.04bA | 6.66 ± 0.03cA | 6.45 ± 0.05dA |
T | 7.04 ± 0.05aB | 6.88 ± 0.07bB | 6.54 ± 0.08cB | 6.30 ± 0.09dB |
Sweetness | ||||
C | 7.05 ± 0.02aA | 6.94 ± 0.02bB | 6.79 ± 0.04cB | 6.38 ± 0.03dB |
T | 7.09 ± 0.03aA | 7.01 ± 0.03aA | 6.84 ± 0.03bA | 6.58 ± 0.02cA |
Overall acceptability | ||||
C | 6.97 ± 0.04aA | 6.87 ± 0.04bA | 6.57 ± 0.03cB | 6.19 ± 0.03bB |
T | 7.1 ± 0.04aA | 6.91 ± 0.03bA | 6.65 ± 0.02cA | 6.23 ± 0.03dA |
*Mean with different superscripts row wise (small letters) and column wise (capital letters) differ significantly (p < 0.05); n = 6; Control: 100% chakka; T1: with 20% carrot pulp. |