Partial Least Squares Research Articles

Intelligent monitoring of post-processing characteristics in 3D-printed food products: A focus on fermentation process of starch-gluten mixture using NIR and multivariate analysis

The production of three-dimensional (3D)-printed food products requires not only optimal 3D-printing adaptability but also appropriate post-processing characteristics. This study aimed to use near infrared (NIR) spectroscopy to predict the rheological properties of 3D-printed dough, enabling intelligent monitoring of the dough's fermentation process. Utilizing support vector machine (SVM) classification model, the fermentation stages can be classified as under-fermentation, complete fermentation, and over-fermentation. Employing preprocessing methods with Synergy Interval Partial Least Square-Competitive Adaptive Reweighted Sampling (SIPLS-CARS) algorithm, 27, 39, 23, and 27 key wavelengths were filtered from the raw NIR spectral data, corresponding to the prediction of storage modulus (G′), loss modulus (G″), complex viscosity (η∗), and loss factor (tan δ), respectively. Quantitatively, SVM (Support Vector Machine) regression outperformed Partial Least Squares (PLS) with Rc2 values (0.95, 0.94, 0.94) and Rp2 values (0.93, 0.93, 0.94) for G′, G″, and η∗. NIR spectra-based predictive models demonstrated superior performance compared to rheo-fermentation properties models. In summary, these findings show the potential of NIR spectroscopy as a rapid tool for predicting the fermentation progress of 3D-printed doughs.

The Effect of Job Demands and Job Resources on Turnover Intention Mediated by Work Engagement at PT Prinal

This research aims to identify the influence of job demands and job resources on turnover intention, with work engagement as a mediating variable at PT Prinal. The retail industry faces high turnover rates, posing a significant challenge at PT Prinal despite the industry's enormous potential. The study involved 250 employees of PT Prinal, using a non-probability sampling technique. Analysis was conducted with SmartPLS v4.1.0.5, utilizing Partial Least Square (PLS) and Structural Equation Model (SEM) techniques to test the hypotheses. The findings show that greater job expectations dramatically reduce employee engagement and increase the likelihood of leaving the position. The job demands at PT Prinal are generally counterproductive, as demonstrated by how they diminish employee engagement. On the other hand, more excellent job resources considerably raise employee engagement at work and lower the intention to leave. Since job demands and job resources directly affect turnover intention, work engagement plays a role in mediating the interaction between these three variables. These results highlight how crucial it is to manage job expectations and resources to boost work engagement and increase employee retention.

Coconut Testa Flour Sub-Fractions: Correlation Between FTIR Spectral Data and α-Glucosidase Inhibitory Activities

Fourier-transform infrared spectroscopy (FTIR) serves as a rapid analytical technique to characterize food specimens chemically. The purpose of this study was to investigate the potential of FTIR combined with multivariate statistics to detect Alpha-glucosidase (Alpha-glu) inhibitory activities of a non-cereal flour-like coconut testa flour (CTF). CTF of five distinct local cultivars was sequentially extracted with hexane, ethyl acetate (EtOAc), and methanol (MeOH) to assay the Alpha-glu inhibitory activity. FTIR spectra of CTF extracts were obtained within the range of 4000–500 cm−1 and the prominent spectral peaks obtained for both hexane and EtOAc extracts were roughly similar but some additional peaks were observed in EtOAc extracts representing phenolic constituents. The major absorbance peaks found in MeOH extracts were primarily indicative of the occurrence of the hydroxyl group associated with carbohydrates and phenolic compounds. The multivariate predictive models developed using partial least squares (PLS) and orthogonal partial least squares (OPLS) regression analyses indicated a strong correlation between Alpha-glu inhibitory activity and spectral data. Models developed for the spectral regions 3700–2800 cm−1 and 1800–500 cm−1 exerted the highest regression coefficients with the lowest root mean square errors. In OPLS regression analysis, the model obtained with third-derivative spectral data was identified as the best, exhibiting the highest regression coefficients and the lowest root mean square errors. Both PLS and OPLS regression analyses indicated a potential correlation of Alpha-glu inhibitory activity with FTIR spectral regions. Notably, OPLS models offered enhanced interpretability of the model parameters. This study suggests that the application of multivariate regression analysis of FTIR spectral data on coconut-based products could help to detect Alpha-glu inhibitory activities.

The Impact of Consumer Ethics on Intention to Purchase Counterfeit Fashion Products

The issue of counterfeiting has long plagued many industries including fashion, automotive, electronics, and pharmaceutical. Companies face various challenges when it comes to dealing with the issue of counterfeiting, which results in financial implications, loss of brand reputation, and many more. From a consumer perspective, counterfeit is the cheaper option in a market where consumers are spoilt for choice. Advancements in technology have helped manufacturers of counterfeit to increase the similarity and to a certain extent the quality of their products to their genuine counterparts. Thus, the demand for counterfeit products in general kept rising over the years. This research looks at the influence of perceived behavioral control and consumer ethics in influencing consumer intention to purchase counterfeit fashion products. A partial least squares (PLS) approach to structural equation modeling (SEM) was used to analyze the data. Findings showed that there is a negative relationship between perceived behavioral control and intention and a positive relationship between consumer ethics and intention.

Adoption of E-Wallets among Students: Trust and Social Influence in E-Commerce

E-commerce introduces the e-wallet payment transaction system as an innovation that offers convenience and security for consumers in purchasing goods and services online. This research explores the influence of trust in social environments in fostering the intention of students in Central Java, Indonesia, to use e-wallets, utilizing technology theories such as the Technology Acceptance Model (TAM). A quantitative study was conducted through an online survey among students. Purposive sampling was employed as the sampling technique, and data were analyzed using the Partial Least Squares (PLS) approach within Structural Equation Modeling (SEM) analysis. The study found that the factors of trust and social influence significantly impact the intention to use e-wallets, with trust even playing a supportive role in fostering this intention alongside other factors. These findings will assist service providers in enhancing e-wallet services with new innovations tailored to user needs.

The Impact of Student Attitude on Information, Communication, Creation Skills and Student Engagement

The Malaysian education system is guided by the Malaysia Education Blueprint, which drives the overall quality and effectiveness of education in the country. This study provides insights and assists in forming informed guidelines for the Ministry of Education regarding the impact of skill factors. The relationship between the direct effects of these three primary skills and attitudes is supported by Self-Efficacy Theory (SET). The indirect relationship, where attitude mediates the effects of these skills on student engagement, is supported by Self-Determination Theory (SDT) and Social Cognitive Theory (SCT). A questionnaire was distributed to local university students currently in their fourth semester and analyzed using the Structural Equation Model using Partial Least Square (PLS). The structural model results direct relationship between attitude and student engagement and between information skills and attitude. In addition, the mediation analysis shows that the relationship between information skills and student engagement, mediated by student attitude, was significant. Based on these results, it is recommended that the Ministry of Education focus on providing supportive resources, such as learning environments and training for educators, in the upcoming Malaysia Education Blueprint for 2026 to 2036. Additionally, offering Digital Literacy Programs to equip students with information skills can positively influence student attitudes and thereby increase the level of student engagement, which is crucial for improving the academic performance and achievement of Malaysia’s educational system. Future research should include a cross-cultural comparison to reconsider educational theories for Asian countries, as many theories are currently based on Western educational settings and contexts.

Determinants of Paying Zakat Through E-Zakat in Tarakan

Research Originality: The existence of adequate infrastructure, especially in the Tarakan City area, makes the author utilize the integration of the conceptual approaches of the Technology Acceptance Model and the Theory of Planned Research Objectives: This study aims to analyze the influence of variables from the Technology Acceptance Model (TAM) theory and the Theory of Planned Behavior (TPB) on the interest in paying zakat, infaq, and alms through e-zakat.Research Methods: This study uses primary data from the distribution of questionnaires using the purposive sampling method with 111 respondents who are Muslim and domiciled in Tarakan City. The research method is the Structural Equation Model (SEM) Partial Least Square (PLS) approach.Empirical Results: The results obtained that attitudes and behavioral control have a significant influence on the interest in use, while the perception of usefulness, perception of ease, and subjective norms do not have a significant influence on the interest in paying zakat, infaq, and alms through e -zakat.Implications: The community still feels comfortable and easy to pay ZIS directly and the presence of people close to them or the surrounding environment has not fully provided confidence to the Muslim community of Tarakan City to use e-zakat as a payment service.JEL Classification: H71, H76, Q53

Factors Impacting the Use of Donation-Based Crowdfunding Platforms in Saudi Arabia: The Case of Ehsan Platform

This research investigates the factors influencing people's intentions to use donation-based crowdfunding platforms in Saudi Arabia, focusing on the Ehsan platform supervised by the Saudi Data and Artificial Intelligence Authority (SDAIA). Ehsan aims to ensure proper monetary distribution to genuinely needy individuals, thereby reducing financial fraud and corruption. For the platform's success, it is essential that citizens effectively utilize it. Employing the Unified Theory of Acceptance and Use of Technology, the study analyzes the determinants that influence intentions to use Ehsan and actual use behaviors. An empirical study was conducted with 224 Saudi citizens from various age groups, and the research model was validated using the Partial Least Squares (PLS) approach. Findings reveal that performance expectancy and facilitating conditions significantly impact intentions to use Ehsan, while effort expectancy and social influence do not. The results indicate a positive association between participants' intentions to use Ehsan and their actual usage. This study enhances the understanding of factors affecting donation-based crowdfunding platforms in Saudi Arabia and offers practical insights for platform developers and decision-makers to consider in their design and modifications.

Unlocking HR’s Hidden Treasure on Innovation Capability: The Mediating Role of Knowledge Management Capacity

Effective innovation is possible when one can innovate. Malaysia has serious problems due to a lack of R&D and innovation skills. Furthermore, small and medium-sized firms (SMEs) in developing nations have received less attention in past research on Malaysian innovation. A quantitative strategy is employed to meet the study's goals. A survey approach is used to collect data from the owners, chief executives, and management of ICT SMEs. The research collected and analyzed a total of 200 respondents. The research models and hypotheses were assessed using the Partial Least Squares (PLS) technique. This research comprises the difference between Innovative Centric Human Resource Management (ICHRM) (Acquisition; Development; Egalitarian; Collaborative; Documentation, Information System) and Innovation Capability (Radical; Incremental), also the relationship between ICHRM and Innovation Capability, using Knowledge Management Capacity as a mediator. Based on the findings of this study, the majority of HRM practices are suitable for SMEs' incremental innovation capabilities. This study also presents empirical evidence of the impact of ICHRM on knowledge management capacity on innovation, focusing on the impact of each dimension in ICHRM components that were linked to organizational processes, with a focus on the influence on incremental and radical innovation.

Understanding Gen Z’s Online Purchase Behavior through their Hedonic and Utilitarian Motivation

The COVID-19 pandemic ushered in lockdowns, limiting consumer movement to their homes and preventing consumers from shopping in brick-and-mortar stores. Consequently, consumers were reliant on e-commerce to manage their daily purchases of goods and services. This exacerbated the use of e-commerce and fast-tracked the growth of the digital economy. In this study, Gen Z’s consumer motivation is investigated using hedonic and utilitarian motivation and the Unified Theory of Acceptance and Use of Technology (UTAUT). Consumer motivation remains a crucial part of consumer behavior studies given its prominence in influencing consumer action, consumer decision-making and preferences. In particular, the effects of hedonic and utilitarian motivation on online purchase behavior were investigated. This study utilized a quantitative approach through the deployment of a survey questionnaire online. The data from 156 respondents was analyzed using SmartPLS 4.0 utilizing Partial-Least Squares Structural Equation Modelling. Approximately 44% of the respondents started shopping online for the first time during the pandemic (i.e., 2020 and 2022). The results indicated that the Gen Z respondents were motivated by utilitarian and hedonic motivation when shopping online, in particular by Idea Shopping motivation and Efficiency motivation. Additionally, Social Influence and Facilitating Conditions were significant factors in influencing Behavioural Intention, and Behavioural Intention influenced Purchase Behaviour. The evidence suggests that the respondents were not novice online shoppers but rather experienced online shoppers.

Combining NIR spectroscopy with chemometrics for discriminating naturally ripened banana and calcium carbide ripened banana

Calcium carbide is prohibited as a fruit ripening agent in many countries due to its harmful effects. Current methods for detecting calcium carbide in fruit involve time-consuming and destructive chemical analysis techniques, necessitating the need for non-destructive and rapid detection techniques. This study combined near infrared (NIR) spectroscopy with chemometrics to detect two banana varieties ripened with calcium carbide in different forms when they are peeled or unpeeled. Sixteen linear discriminant analysis (LDA) models were developed with high average classification accuracies for classifying banana based on the mode used to ripen banana, type of carbide treatment and the duration of soaking banana in carbide solution. Banana colour was predicted with partial least squared regression (PLSR) models with R2CV > 0.74, RMSECV and <5.4 and RPD close to 3. NIR coupled with chemometrics has good potential as a technique for detecting carbide ripened banana even if the banana is peeled or not.

PENGARUH BRAND IMAGE DAN HARGA TERHADAP KEPUASAN KONSUMEN AQUA MELALUI KEPUTUSAN PEMBELIAN DI SUPERMARKET HARI–HARI KOTA DUMAI RIAU

This research aims to determine the influence of brand image and price on consumer satisfaction at Hari - Hari Dumai City Riau Supermarket through intervening variables in purchasing decisions. This research uses a quantitative method with an explanatory research type. The sample in this study was taken as many as 100 people using accidental sampling technique. Hypothesis testing was carried out using a Structural Equation Model (SEM) approach based on Partial Least Square (PLS) using the SmartPLS version 3 application to test the relationship between variables. The research results show that brand image has an insignificant positive influence on Aqua purchasing decisions at the Hari - Hari Dumai City Riau Supermarket. Price and brand image have significant positive influence on Aqua purchasing decisions at the Hari - Hari Dumai City Riau Supermarket. Price does not have a significant effect on Aqua consumer satisfaction at Hari - Hari Dumai City Riau Supermarket. Purchasing decisions have a significant effect on Aqua consumer satisfaction at Hari - Hari Dumai City Riau Supermarket. Brand image has an insignificant indirect effect on Aqua consumer satisfaction through purchasing decisions at the Hari - Hari Dumai City Riau Supermarket. Price has an indirect effect on Aqua consumer satisfaction through purchasing decisions at the Hari - Hari Dumai City Riau Supermarket.

PENGARUH KUALITAS PRODUK DAN KEPERCAYAAN PELANGGAN TERHADAP KEPUTUSAN PEMBELIAN MOTOR MELALUI KEPUASAN PELANGGAN PADA PT. SURYA GEMILANG MOTOR SURABAYA

The purpose of this research is to test and analyze the influence of product quality, customer trust on purchasing decisions through customer satisfaction at PT. Surya Gemilang Motor, West Surabaya. Sampling method using Purposive Sampling. The scope of this research is only on variables related to product quality, customer trust which directly or indirectly influence purchasing decisions through customer satisfaction. The approach of this research is a quantitative approach method. The data analysis technique for this research uses the Partial Least Square (PLS) method with the SmartPLS 4.0 application program which includes the design process of the Inner Structural Model, Outer Structural Model and Hypothesis Testing (Resampling Bootstrapping). The research results found that there was no direct effect of product quality on customer satisfaction at PT. Surya Gemilang Motor Surabaya. There is a direct influence of Customer Trust on Customer Satisfaction at PT. Surya Gemilang Motor, West Surabaya. There is a direct effect of product quality on purchasing decisions at PT. Surya Gemilang Motor, West Surabaya. There is a direct influence of Customer Trust on Purchasing Decisions at PT. Surya Gemilang Motor, West Surabaya. There is a direct effect of purchasing decisions on customer satisfaction at PT. Surya Gemilang Motor, West Surabaya. There is no effect of Product Quality indirectly on Purchase Decisions through Customer Satisfaction at PT. Surya Gemilang Motor, West Surabaya. There is an indirect effect of Customer Trust on Purchasing Decisions through Customer Satisfaction at PT. Surya Gemilang Motor.

From Play to Pay: Exploring Imaginal and Emotional Virtual Item Retail Experiences in Online Game Environment

In online gaming, selling virtual game items is a growing phenomenon designed to engage players. Despite growing adoption, little remains known regarding the gamer’s perceived virtual item retail experiences, particularly in the online game retail context. Based on the hedonic consumption perspectives, we examined the role of escapism, role projection, fantasy, enjoyment, emotional involvement, and arousal in developing positive attitudes toward in-game virtual items, which eventually enhanced behavioral outcomes such as game store loyalty, willingness to pay (WTP) and word-of-mouth (WOM). We collected survey data from 281 online gamers and analyzed the data using partial least squares (PLS)-based structural equation modeling. The results indicate that fantasy, escapism, enjoyment, and arousal significantly enhance gamers’ favorable attitudes toward virtual item experiences. Moreover, the findings indicate the positive influence of attitude on players’ WTP, loyalty to retailers, and positive WOM. This study is the first to explore imaginal and emotional experiences as antecedents that derive gamers’ favorable attitudes toward in-game item retail experiences and their subsequent influence on gamers’ consumption behavior. Online game retailers may benefit from focusing on virtual item experiences to elevate premium gamers’ retail attitudes and their ensuing consumption behaviors.

The Impact of Compensation and Work Motivation on Job Satisfaction of PT Toraya Cipta Gemilang Production Employees in Mojokerto Regency

The research aims to determine the influence of compensation and work motivation on job satisfaction of production employees of PT. Toraya Cipta Gemilang, Mojokerto Regency. The method used in this research is a quantitative method. Sampling used non-probability sampling techniques using saturated sampling. A saturated sample was used in this study because the entire population was used. The population in this study was 86 people, so the sample used in this study was 86 people. The data analysis technique used in this research uses Partial least squares (PLS). The results of the research show that the independent variable (X1) compensation contributes to the dependent variable job satisfaction (Y) and the independent variable work motivation (X2) contributes to the dependent variable job satisfaction (Y) of production employees of PT. Toraya Cipta Gemilang in Mojokerto Regency. Providing competitive compensation has a direct impact on individual happiness in the organization, greater compensation means happier employees and the provision of motivation carried out by various parties has an impact on employee job satisfaction in the company, thus having a positive impact on the company as well.

Multi-Sensor Soil Probe and Machine Learning Modeling for Predicting Soil Properties

We present a data-driven, in situ proximal multi-sensor digital soil mapping approach to develop digital twins for multiple agricultural fields. A novel Digital Soil CoreTM (DSC) Probe was engineered that contains seven sensors, each of a distinct modality, including sleeve friction, tip force, dielectric permittivity, electrical resistivity, soil imagery, acoustics, and visible and near-infrared spectroscopy. The DSC System integrates the DSC Probe, DSC software (v2023.10), and deployment equipment components to sense soil characteristics at a high vertical spatial resolution (mm scale) along in situ soil profiles up to a depth of 120 cm in about 60 s. The DSC Probe in situ proximal data are harmonized into a data cube providing vertical high-density knowledge associated with physical–chemical–biological soil conditions. In contrast, conventional ex situ soil samples derived from soil cores, soil pits, or surface samples analyzed using laboratory and other methods are bound by a substantially coarser spatial resolution and multiple compounding errors. Our objective was to investigate the effects of the mismatched scale between high-resolution in situ proximal sensor data and coarser-resolution ex situ soil laboratory measurements to develop soil prediction models. Our study was conducted in central California soil in almond orchards. We collected DSC sensor data and spatially co-located soil cores that were sliced into narrow layers for laboratory-based soil measurements. Partial Least Squares Regression (PLSR) cross-validation was used to compare the results of testing four data integration methods. Method A reduced the high-resolution sensor data to discrete values paired with layer-based soil laboratory measurements. Method B used stochastic distributions of sensor data paired with layer-based soil laboratory measurements. Method C allocated the same soil analytical data to each one of the high-resolution multi-sensor data within a soil layer. Method D linked the high-density multi-sensor soil data directly to crop responses (crop performance and behavior metrics), bypassing costly laboratory soil analysis. Overall, the soil models derived from Method C outperformed Methods A and B. Soil predictions derived using Method D were the most cost-effective for directly assessing soil–crop relationships, making this method well suited for industrial-scale precision agriculture applications.

Organizational support as a moderator on academic entrepreneurship performance in state universities

Engaging in academic entrepreneurship enhances the potential for academics to generate scientific advancements, capitalize on findings, and create impactful contributions. This study examines how organizational support affects the relationship between entrepreneurial orientation and role integration on academic entrepreneurship performance at state universities in Indonesia. The population includes lecturers from eleven state universities with research results that can be commercialized. A purposive sampling method selected 330 respondents, focusing on science and social sciences lecturers with research at a technological readiness level of 6 to 9, demonstrating pilot stages and commercialization readiness based on national assessments. Data analysis used variance-based structural equation modeling and partial least squares (PLS). The findings indicate that entrepreneurial orientation significantly affects academic entrepreneurship performance (p &amp;amp;lt; 0.05), and role integration significantly affects academic entrepreneurship performance (p &amp;amp;lt; 0.05). Additionally, organizational support moderates the relationship between entrepreneurial orientation and academic entrepreneurship performance (p &amp;amp;lt; 0.05), and organizational support moderates the relationship between role integration and academic entrepreneurship performance (p &amp;amp;lt; 0.05). AcknowledgmentsWe would like to thank the lecturers at eleven state universities in Indonesia who contributed to this investigation. Their assistance allowed the authors to obtain data that met the study criteria and generate accurate and exact data analysis per the research goals.

Influence of service, innovation, integrity on taxpayer compliance with education as a moderating variable at the UPT PPD/Samsat Karimun Office

Purpose: This study analyzes the influence of service, innovation, and integrity on taxpayer compliance with education as a moderating variable in the UPT PPD/Samsat Karimun office. Methodology: This study used a quantitative and descriptive approach with a population of all 2-Wheeled Motor Vehicle Taxpayers, namely 64,003 people, to obtain a sample of 156 respondents. This study used primary data with data sources, distributing Google form questionnaires, and going directly to the field. The data analysis technique used is the partial least squares (PLS) approach, which is a structural equation model (Structural Equation Modeling or SEM) based on components or variants, SmartPLS 4.0. Results: Based on the results of this study, it can be concluded that 1) service quality has a significant positive effect on taxpayer compliance, with an original sample value (O) of 0.305 and T-statistic of 2.235. 2) Innovation does not have a significant effect on taxpayer compliance, with an original sample value (O) of 0.252 and T-statistic of 0.870. 3) Integrity has a significant positive effect on taxpayer compliance, with an original sample value (O) of 0.370 and T-statistic of 2.082. 4) Service does not have a significant effect on tax education, with an original sample (O) value of 0.043 and T-statistic of 0.197. 5) Innovation has a significant positive effect on tax education, with an original sample (O) value of 0.313 and T-statistic of 1.984. 6) Integrity does not have a significant effect on tax education, with an original sample (O) value of 0.529 and T-statistic of 1.916. 7) Service has a significant positive effect on taxpayer compliance through education, with an original sample (O) value of 0.308 and T-statistic of 2.091. 8) Innovation does not have a significant effect on taxpayer compliance through education, with an original sample (O) value of 0.059 and T-statistic of 0.367. 9) Integrity does not have a significant effect on taxpayer compliance through education, with an original sample (O) value of 0.100 and T-statistic of 0.510. 10) Education has a significant positive influence on taxpayer compliance, with an original sample value (O) of 0.419 and T-statistic of 1.976.

Raman spectroscopy coupled with the PLSR model: A rapid method for analyzing gamma-oryzanol content in rice bran oil

Rice bran oil (RBO) is widely used in food, nutraceutical, and cosmetic industries, due to its γ-oryzanol content, a key quality indicator. This study developed a rapid, non-destructive method for quantifying γ-oryzanol in RBO using Raman spectroscopy combined with partial least squares regression (PLSR). The optimal PLSR model, based on orthogonal signal correction (OSC)-pretreated data of Raman spectra from 800 to 1800 cm−1, demonstrated high accuracy with a strong R2-Pearson correlation coefficient of 0.9827 and low root mean square error of prediction (RMSEP) of 0.5314. Principal component analysis (PCA) of OSC-pretreated data showed improved sample grouping by concentration of γ-oryzanol compared to untreated data. Additionally, Bland-Altman plots comparing results from Raman and HPLC methods showed random scatter within ±2 SD of the mean difference, confirming the method's reliability. This study indicates that Raman spectroscopy can serve as a reliable method for determining γ-oryzanol content in RBO products within the related industries.

Ambient temperature-dependent variations in bitter compounds and microbial dynamics in traditional huangjiu fermentation

Huangjiu, a traditional Chinese fermented wine, often develops excessive bitterness under semi-open fermentation conditions. This study examines the impact of ambient temperature on bitterness and microbial composition during Huangjiu fermentation. Using high−performance liquid chromatography (HPLC), gas chromatography−olfactometry (GC−O), and metagenomic sequencing, we identified seven bitter amino acids and eleven bitter volatile substances. Our results indicate that lower ambient temperatures (8.21 ± 0.69 °C to 9.88 ± 0.83 °C) suppress the accumulation of bitter substances, with the total concentration of bitter amino acids reduced by up to 3.94% compared to higher temperatures (16.36 ± 0.97 °C to 4.75 ± 0.74 °C). Partial least squares (PLS) regression established correlations between electronic tongue bitterness values and the presence of bitter compounds, revealing that higher temperatures promote the formation of specific compounds, such as benzaldehyde, furfural, ethyl nonanoate, and phenylalanine, thereby enhancing bitterness perception. Metabolic pathway analysis highlighted the enzymatic processes and intermediates involved in the biosynthesis of these bitter compounds. Correlation analysis showed that Saccharomyces and Lactobacillus are closely linked to the synthesis of bitter amino acids, while Aspergillus and Rhizopus contribute to the formation of furfural and ethyl octanoate. This study offers a theoretical foundation for reducing flavor variations between Huangjiu batches and provides practical guidance for optimizing the traditional fermentation process.