Households are responsible for up to 72% of global greenhouse gas emissions through everyday choices in food, transport, housing, and energy — yet most consumers who express concern about climate change still fail to translate those attitudes into action (Newell et al., 2021; Yin et al., 2025). This article examines the science behind consumer carbon habits, the AI technologies that can predict and reshape them, and — critically — how to measure whether change is actually happening. Drawing on 28 peer-reviewed studies, it builds a comprehensive picture of where artificial intelligence fits into the behaviour-change puzzle, from big-data nudges and personalised carbon tracking to gamification and real-time feedback.
Introduction
Climate change represents one of the most pressing challenges of our time, driven in large part by the cumulative carbon emissions arising from everyday consumer behaviour. With global CO₂ emissions reaching approximately 38.81 Gt CO₂ in 2020 and an average per capita carbon footprint of around 4.47–4.72 metric tonnes (Raghavasamy.M et al., 2024; Brkljač & Lukić, 2023), the urgency to transform consumption patterns has never been greater. Households are responsible for a substantial share of greenhouse gas (GHG) emissions — by some estimates up to 72% — through their consumer choices in food, transport, housing, and energy use (Newell et al., 2021). Yet despite growing awareness, a persistent “attitude-behaviour gap” means that many consumers who express concern about the environment fail to translate those attitudes into meaningful action (Yin et al., 2025; Steiner et al., 2017).
This article explores the current landscape of consumer habits related to climate change, examines the barriers to sustainable behaviour, and outlines how recent advances in artificial intelligence (AI), big data analytics, and related technologies offer a powerful toolkit for predicting and reshaping consumer behaviour to reduce CO₂ footprints. Critically, it also addresses the essential question of how we can measure the change in users’ behaviours — a prerequisite for any effective intervention strategy.
Understanding Consumer Habits and Their Climate Impact
The Carbon Footprint of Everyday Life
The carbon footprint of individuals and households is shaped by a complex web of daily decisions — from the food we eat and the products we buy to how we heat our homes and how we travel. Climate change is driven by the rapid advancement of the global economy, persistent population growth, and the resulting surge in energy consumption and carbon emissions (Du et al., 2024). Within the household sector alone, energy consumption accounts for more than 30% of total global energy use, and the continued growth of household energy consumption and associated carbon emissions threatens both economic and environmental sustainability (Wang et al., 2024). Buildings consume approximately 40% of the world’s energy and account for 30% of CO₂ emissions (Youssef & Zeqiri, 2022). Transportation, particularly aviation and personal vehicle use, is another major contributor, with the transport sector depending primarily on fossil fuels and serving as a significant driver of climate change (Wu et al., 2024).
Consumer behaviour in these domains is shaped by deeply ingrained habits, economic incentives, social norms, and psychological biases. Research has shown that socio-demographic factors such as gender, age, education, income, and region of provenance all influence how consumers respond to environmental information and sustainability labels (Rondoni & Grasso, 2021). Furthermore, evolved psychological biases — such as price sensitivity and status quo bias — can make ecologically oriented choices more difficult for certain consumer segments (Steiner et al., 2017).
The Attitude-Behaviour Gap
A central challenge in promoting sustainable consumption is the well-documented gap between environmental attitudes and actual purchasing behaviour. While the concept of green consumption is gradually being accepted by the public, actual purchasing behaviours are often constrained by factors such as price, convenience, and brand influence (Yin et al., 2025). For example, studies have found that while Finnish consumers were familiar with the term “product carbon footprint,” very few could describe it accurately, and most failed to connect it with concepts such as climate change or greenhouse gases (Rondoni & Grasso, 2021). Similarly, in the tourism sector, individual concern for the environment tends to be unrelated to holiday behaviour, suggesting either an unawareness of tourism’s environmental impacts or a state of cognitive dissonance in which consumers justify enjoyment even when aware of environmental damage (Holden et al., 2022).
This gap is not merely a matter of individual willpower. Structural factors — including the availability and affordability of sustainable alternatives, the design of choice architectures, and the influence of social networks — play a decisive role in shaping behaviour (Newell et al., 2021). As Newell et al. argue, relying on conscientious individuals to “do their bit” will never be sufficient without substantial shifts in the behaviour of high-consuming groups and systemic changes in policy, infrastructure, and service provision.
The Role of AI and Big Data in Predicting Consumer Behaviour
Big Data Analytics for Understanding Consumption Patterns
The explosion of digital data — from smart meters and IoT devices to social media activity and e-commerce transactions — has created unprecedented opportunities to understand and predict consumer behaviour at scale. Big data analytics (BDA) enables businesses and policymakers to study individual energy consumption, low-carbon emission transportation, waste resource reuse, and recycling at both micro and macro levels (Chandra & Verma, 2021). Predictive BDA, augmented with machine learning tools, has been shown to improve efficiency in predicting consumer preferences for eco-friendly products and services, such as sustainable hotels and green supply chain products (Chandra & Verma, 2021).
In the household energy domain, smart meter data and non-intrusive load monitoring (NILM) technologies can provide fine-grained consumption profiles per appliance type, enabling both utility companies and consumers to identify energy-saving opportunities (Trakadas et al., 2022). Neural network frameworks have been applied to analyse residential electricity consumption habits and predict energy loads, offering a foundation for personalised recommendations (Wang et al., 2024). These data-driven approaches not only provide a scientific basis for decision-makers but also reveal potential patterns and influencing factors of carbon emissions, offering innovative tools for formulating more effective carbon reduction strategies (Du et al., 2024).
AI-Powered Prediction and Personalisation
Artificial intelligence — encompassing machine learning, deep learning, and generative AI — has emerged as a transformative force for predicting and influencing consumer behaviour. AI algorithms can analyse vast amounts of financial, environmental, and behavioural data to identify patterns, forecast trends, and offer tailored strategies for minimising carbon footprints across various sectors (Parhamfar et al., 2024; WEN, 2024). In the context of sustainable consumption, AI enables the automation of data collection, analysis, and segmentation, facilitating predictive analytics that allow businesses to forecast consumer preferences and purchasing patterns (Jaafar et al., 2024). AI techniques are crucial in reshaping consumer-behaviour analysis by leveraging data insights, and deep learning models have been developed to efficiently classify complex consumer-behaviour variants with high precision, outperforming traditional machine-learning models (Ola et al., 2024).
For example, AI-powered tools can provide individuals and organisations with personalised carbon footprint tracking, offering insights into their emissions and suggesting tailored strategies for reducing their carbon impact (Parhamfar et al., 2024). CO₂ AI platforms utilise machine learning and advanced analytics to help organisations measure, analyse, and strategise around carbon emissions, providing precise values and actionable recommendations (Raghavasamy.M et al., 2024). Furthermore, by analysing vast amounts of carbon footprint data using big data and combining AI models and algorithms, it is possible to more accurately assess households’ carbon emissions and predict future carbon footprint change trends (Du et al., 2024).
AI-Driven Approaches to Changing Consumer Behaviour
Nudging and Choice Architecture
One of the most promising applications of AI in behaviour change is the concept of “nudging” — subtly altering the architecture of choices to influence people’s behaviour without restricting their freedom (Yin et al., 2025; Newell et al., 2021). As popularised by Thaler and Sunstein, a nudge is “any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives.” Digital nudges, powered by AI, can effectively steer consumers towards more sustainable choices during online purchases, enhancing the overall environmental impact of e-commerce, especially among those concerned about climate change (Yin et al., 2025). In the food waste domain, nudge theory has been applied through AI-driven tools such as smart refrigerators and AI-powered apps that track food expiration dates and suggest meal plans, reinforcing responsible food consumption behaviours (Clark et al., 2025).
However, the use of AI for nudging is not without controversy. Corporations have been found to engage in “hyper-nudging,” where consumer behaviour is extensively studied and manoeuvred through AI to lead to hyper-consumerism and heavy consumption (Dubey & Alam, 2024). Data from websites, social media, applications, and e-commerce platforms contribute towards prediction and evaluation of consumer behaviour using AI technologies, and while this can be harnessed for sustainability, it can also be exploited for unsustainable ends (Dubey & Alam, 2024). The ethical implications of AI-driven nudging — including concerns about manipulation, privacy, and the potential for rebound effects — must be carefully addressed (Dubey & Alam, 2024; Newell et al., 2021).
In the smart city context, AI-backed virtual assistants and smart home devices have the potential to nudge consumers towards energy-saving behaviour within their homes. For instance, AI systems can learn individual preferences and tweak the environment to suit them while simultaneously encouraging energy-saving actions (Zamponi & Barbierato, 2022). Yet, as Zamponi and Barbierato note, such technologies require consumers to actively choose to adopt them, and not all consumers may be interested in acquiring energy-saving devices.
Generative AI and the “Dual G.R.E.E.N Circle” Framework
Recent advances in generative AI — which goes beyond traditional predictive models to create new content such as text, images, and interactive experiences — offer novel pathways for promoting green consumption. Yin et al. propose the “Dual G.R.E.E.N Circle” framework, which delineates both the intrinsic mechanisms and practical frameworks for embedding generative AI to facilitate green consumption (Yin et al., 2025). The inner circle describes mechanisms including Guiding Awareness, Rewarding Behaviour, Empowering Accessibility, Enhancing Experience, and Nudging Strategy, while the outer circle outlines practical applications such as Generational Influence, Result Visualisation, Enabling Creativity, Entertaining Education, and Narrative Construction.
This framework addresses the attitude-behaviour gap by leveraging generative AI to design personalised, engaging, and interactive learning environments that make environmental education more accessible and actionable (Yin et al., 2025). For example, AI-enhanced gamification can simulate sustainability challenges and solutions, stimulating creativity and innovation in educational settings while providing cost-effective experiential learning. By tailoring content to users’ interests and behavioural patterns, generative AI can bridge the gap between environmental awareness and concrete action.
Generative AI for Environmental Information and Green Purchasing
Empirical research supports the effectiveness of generative AI in promoting sustainable consumer behaviour. Foroughi et al. demonstrate that interactivity, responsiveness, knowledge acquisition and application, environmental concern, and ascription of responsibility are key predictors of generative AI use for environmental information (Foroughi et al., 2025). Furthermore, the use of generative AI for environmental education boosts awareness and translates into tangible actions that reflect personal environmental responsibility and sustainable consumption habits. The technology’s role in empowering users to comprehend and measure their environmental footprint is particularly significant, as it enables individuals to make informed decisions that align with their values.
Generative AI can also be deployed to raise awareness about environmental sustainability through AI-generated content, and to guide consumers toward eco-friendly purchasing decisions (Foroughi et al., 2025). However, research also cautions that ineffective utilisation of AI could potentially decrease consumer intentions towards green consumption, particularly when compared with human recommenders — a phenomenon known as the “machine recommendation” effect (Yin et al., 2025). This underscores the importance of designing AI systems that are trustworthy, transparent, and aligned with user expectations.
Personalised Carbon Footprint Tracking and Feedback
A critical application of AI for behaviour change is the development of personalised carbon footprint calculators and tracking tools. Traditional online tools for calculating carbon footprints require manual data entry, making them inefficient and cumbersome (Raghavasamy.M et al., 2024). AI-powered platforms can automate data collection and analysis, providing users with real-time, personalised feedback on their emissions and actionable recommendations for reduction (Raghavasamy.M et al., 2024; Parhamfar et al., 2024). IoT-enabled devices such as smart meters and appliances provide real-time energy usage data, empowering consumers to make informed decisions about their energy consumption habits (WEN, 2024).
Emission trackers and mobile apps help consumers trace their carbon footprint and recycling habits, and climate-friendly technological innovations can help consumers engage in environmentally friendly behaviours (Rayburn et al., 2025). For climate-conscious consumers, the first eco-friendly purchase often starts a habit of such behaviour, suggesting that timely, personalised feedback can catalyse a virtuous cycle of sustainable action. By combining AI-driven analytics with behavioural insights, companies can design interventions that are not only informative but also motivating and habit-forming.
Measuring the Change in Users’ Behaviours
A fundamental prerequisite for any AI-driven behaviour change initiative is the ability to measure whether and to what extent users’ behaviours have actually changed. Without robust measurement, it is impossible to evaluate the effectiveness of interventions, iterate on strategies, or demonstrate impact to stakeholders. This section outlines the key approaches, metrics, frameworks, and technologies for measuring behavioural change in the context of reducing consumers’ CO₂ footprints.
Defining Behavioural Metrics and Key Performance Indicators
The first step in measuring behaviour change is to define clear, quantifiable metrics that capture the dimensions of behaviour most relevant to carbon reduction. In the household energy domain, metrics such as kilowatt-hours (kWh) consumed per household, peak-to-off-peak consumption ratios, and appliance-level energy profiles serve as direct indicators of behavioural shifts (Wang et al., 2024; Trakadas et al., 2022). Smart meter data and NILM technologies enable the disaggregation of total household energy consumption into appliance-specific profiles, allowing for the identification of changes in usage patterns over time. Similarly, in the transportation domain, metrics such as vehicle kilometres travelled, modal share (e.g., proportion of trips by public transport, cycling, or walking versus private car), and fuel consumption per trip can be tracked using GPS data, mobility apps, and connected vehicle platforms (Wu et al., 2024).
For food-related behaviours, AI-driven food waste management systems in the hospitality industry have demonstrated the use of metrics such as kilograms of food waste per meal served, food cost savings, and waste categorisation data to quantify behavioural change (Clark et al., 2025). In the broader context of green purchasing, metrics such as the proportion of eco-labelled products in a consumer’s shopping basket, frequency of sustainable product purchases, and changes in brand preferences over time can be derived from e-commerce transaction data and loyalty programme records (Foroughi et al., 2025; Chandra & Verma, 2021).
AI and Machine Learning for Behavioural Pattern Detection and Change Quantification
Machine learning algorithms are central to the task of detecting and quantifying behavioural change from complex, high-dimensional data. Predictive analytics models can establish baseline behavioural profiles for individual users and then detect deviations from these baselines that indicate meaningful change (Chandra & Verma, 2021; Du et al., 2024). For example, neural network frameworks applied to residential electricity consumption data can model habitual energy use patterns and flag statistically significant reductions in consumption following an intervention (Wang et al., 2024). Similarly, machine learning models trained on purchasing data can identify shifts in consumer preferences towards eco-friendly products, quantifying the magnitude and persistence of these shifts over time (Foroughi et al., 2025; Chandra & Verma, 2021).
An emergent group of information systems combine AI techniques such as machine learning with behavioural analytics to play the role of a coach or co-regulator, supporting users in self-improvement (Cranefield et al., 2022). Digital Productivity Assistants (DPAs), for instance, apply machine learning to data created by users’ activity to coach users in improving their behaviours, providing an overview of people’s patterns and actionable advice for how individuals can change their behaviour. These systems apply a range of behavioural techniques, such as comparative metrics and evidence-based justifications for particular changes, in order to nudge users towards a desired behaviour. The measurement of change in this context involves tracking the adoption and persistence of recommended behaviours over consecutive time frames, using the AI system’s own analytics to quantify progress.
In the context of sustainability, AI-powered carbon footprint tracking platforms can measure changes in users’ emissions over time by integrating data from multiple sources — energy bills, transport logs, purchasing records, and IoT sensors — and computing personalised carbon footprint trajectories (Raghavasamy.M et al., 2024; Du et al., 2024; Parhamfar et al., 2024). By comparing current emissions against historical baselines and against personalised reduction targets, these platforms can provide users and organisations with clear, quantitative evidence of behavioural change.
Behavioural Frameworks and Theoretical Models for Measurement
Robust measurement of behaviour change requires grounding in established theoretical frameworks. The Theory of Planned Behaviour (TPB), the Value-Belief-Norm (VBN) theory, and the Elaboration Likelihood Model (ELM) provide structured approaches for understanding and measuring the cognitive, attitudinal, and normative antecedents of behaviour change (Foroughi et al., 2025; Clark et al., 2025). Foroughi et al. integrate these theories to pinpoint and prioritise determinants of generative AI usage for environmental information and green purchasing behaviour, using a hybrid methodology that blends partial least squares (PLS) with artificial neural networks (ANN) to analyse data from 467 participants (Foroughi et al., 2025). The PLS outcomes indicate that interactivity, responsiveness, knowledge acquisition and application, environmental concern, and ascription of responsibility are key predictors, while the ANN analysis offers a unique perspective and discloses variations in the hierarchy of these predictors. This hybrid approach exemplifies how AI can be used not only to drive behaviour change but also to measure and model the factors that underlie it.
The “Dual G.R.E.E.N Circle” framework proposed by Yin et al. also provides a structured basis for measurement, delineating both the mechanisms and the practical applications through which generative AI can promote green consumption (Yin et al., 2025). Each element of this framework can be operationalised as a measurable construct, enabling researchers and practitioners to assess the effectiveness of specific AI-driven interventions along multiple dimensions.
From a broader behavioural science perspective, Newell et al. advocate for a move away from linear and “shallow” understandings of behaviour change towards a “deep,” contextualised and dynamic view of scaling as a transformative process of multiple feedbacks and learning loops between individuals and systems (Newell et al., 2021). This implies that measurement should capture not only discrete behavioural outcomes (e.g., kWh saved, kg CO₂ reduced) but also the dynamic processes of attitude formation, social norm diffusion, and systemic feedback that sustain long-term change.
Longitudinal Tracking and Habit Formation Metrics
Measuring behaviour change is not a one-time event but a longitudinal process. The persistence and durability of behavioural shifts are as important as their initial magnitude. Longitudinal tracking — monitoring users’ behaviours over weeks, months, and years — is essential for distinguishing transient responses from lasting habit formation (Rayburn et al., 2025; Ghani et al., 2023). For climate-conscious consumers, the first eco-friendly purchase often starts a habit of such behaviour, suggesting that early measurement of adoption can predict long-term engagement (Rayburn et al., 2025).
AI systems are well suited to longitudinal measurement because they can continuously collect and analyse data, adapting their models as user behaviour evolves. DPAs, for example, analyse user behaviour over consecutive time frames to predict the possibility of behavioural change within a following time window and prescribe actions accordingly; long-term usage of such systems may even change the actual behavioural patterns of a user (Cranefield et al., 2022; Raychaudhuri et al., 2021). In the food waste domain, AI systems learn from user behaviour and adjust their recommendations over time, and the measurement of their impact requires tracking waste reduction metrics over extended periods to capture the full trajectory of behavioural change (Clark et al., 2025).
Self-Reported and Survey-Based Measures
While passive, data-driven measurement is increasingly powerful, self-reported measures remain an important complement — particularly for capturing subjective dimensions of behaviour change such as attitudes, motivations, and perceived barriers. Surveys and questionnaires grounded in established psychometric scales (e.g., the New Ecological Paradigm, environmental concern scales, and green purchasing intention scales) can be administered at multiple time points to track changes in consumers’ environmental attitudes and self-reported behaviours (Steiner et al., 2017; Rondoni & Grasso, 2021; Foroughi et al., 2025). Latent class analysis and other segmentation techniques can be employed to identify and characterise distinct consumer segments and to track how individuals move between segments over time in response to interventions (Steiner et al., 2017).
However, self-reported measures are subject to well-known biases, including social desirability bias and recall error (Rondoni & Grasso, 2021). The integration of self-reported data with objective, sensor-derived behavioural data provides a more complete and reliable picture of behaviour change, enabling triangulation and validation of findings (Coman et al., 2024; Trakadas et al., 2022).
Composite and Multi-Dimensional Measurement Frameworks
Given the complexity of consumer behaviour and the multiplicity of domains in which carbon emissions arise, effective measurement requires composite, multi-dimensional frameworks. Du et al. propose a multidimensional and multi-level framework for assessing household carbon footprints, integrating data on energy consumption, transportation, food, and waste at individual, household, and city levels (Du et al., 2024). Such frameworks can serve as the basis for measuring behaviour change across multiple domains simultaneously, providing a holistic view of a user’s carbon trajectory.
In the digital marketing and AI domain, evaluation metrics such as segment homogeneity, engagement rates, conversion rates, and retention rates are used to assess the effectiveness of AI-optimised segmentation and targeting strategies (Khair, 2024). These metrics can be adapted for sustainability applications — for example, measuring the proportion of users who adopt recommended sustainable behaviours, the frequency and duration of engagement with carbon tracking tools, and the net change in estimated carbon footprint over time.
The integration of these diverse measurement approaches — IoT-derived behavioural data, AI-driven pattern detection, theoretical frameworks, experimental designs, longitudinal tracking, self-reported measures, and composite indices — constitutes a robust, multi-layered measurement ecosystem. For an AI company seeking to predict and change users’ behaviour to reduce their CO₂ footprint, this ecosystem provides the evidence base needed to demonstrate impact, refine interventions, and scale successful strategies.
Addressing Challenges and Ethical Considerations
Ethical and Social Implications
The use of AI for behaviour change raises important ethical questions about autonomy, privacy, and equity. AI-driven nudging, if not carefully designed and regulated, can cross the line from gentle persuasion to manipulation (Dubey & Alam, 2024; Newell et al., 2021). The collection and analysis of personal data for behavioural prediction must be governed by robust privacy protections and transparent data practices. Consumer trust is a significant barrier to adopting AI-driven technologies, as many consumers are wary of sharing personal data with AI systems, particularly when it comes to sensitive information about their purchasing habits and consumption patterns (Clark et al., 2025). Moreover, the benefits of AI-driven sustainability interventions must be distributed equitably, avoiding the exacerbation of global inequalities (Dubey & Alam, 2024; Newell et al., 2021).
From a psychological perspective, a major deficiency in the effectiveness of nudging as a tool for behaviour change is that it fails to engage with the attitudes, values, and beliefs underlying individuals’ motivations for taking action (Newell et al., 2021). Social and environmental psychology bring greater cognitive depth to our understanding of how human behavioural responses can be used to promote climate mitigation and adaptation, and AI systems should be designed to complement — not replace — these deeper forms of engagement.
Conclusion
Consumer habits are at the heart of the climate challenge, but they are also a powerful lever for change. The persistent attitude-behaviour gap in green consumption can be addressed through the intelligent application of AI and big data technologies — predicting individual carbon footprints, personalising feedback and recommendations, nudging consumers towards sustainable choices, and engaging them through interactive and educational experiences. Crucially, the ability to measure behavioural change — through IoT-derived data, AI-driven pattern detection, experimental designs, longitudinal tracking, and composite measurement frameworks — is what transforms these interventions from aspirational to evidence-based. Without rigorous measurement, it is impossible to know whether interventions are working, to iterate and improve, or to demonstrate impact at scale. However, these technological interventions must be designed with careful attention to ethical considerations, transparency, and equity, and must be embedded within a broader ecosystem of systemic change involving governments, corporations, and civil society. By harnessing the full potential of AI to understand, transform, and measure consumer behaviour, we can accelerate the transition to a sustainable, low-carbon future.
This is precisely the territory SAVE is built to navigate. Rather than asking users to audit their habits manually or engage with abstract carbon totals, SAVE learns individual patterns on-device and delivers well-timed, context-aware suggestions — the kind of personalised nudge the research above identifies as most effective. Measurement is built in from the start: every interaction is an opportunity to track whether habits are shifting, and by how much.
Sources
- Exploring Consumer Perceptions in Online Shopping for Sustainable Economic Development (2024)
- Green Employer Branding and Company Attractiveness for Customers in the Digital Age — Brkljač & Lukić (2023)
- Big Data and Sustainable Consumption: A Review and Research Agenda — Chandra & Verma (2021)
- AI-driven food waste management strategies in the hospitality industry — Clark et al. (2025)
- Smart Solutions for Diet-Related Disease Management — Coman et al. (2024)
- Partnering with AI: digital productivity assistants — Cranefield et al. (2022)
- A review of household carbon footprint: measurement methods, evolution and emissions assessment — Du et al. (2024)
- Corporations greening strategies: Overlooking the negative implications of AI’s contributions? — Dubey & Alam (2024)
- Determinants of Generative AI in Promoting Green Purchasing Behavior — Foroughi et al. (2025)
- Current Approaches of AI in Leading Behavioural Change — Ghani et al. (2023)
- The Future of Tourism in the Anthropocene — Holden et al. (2022)
- AI-Powered Digital Marketing: Elevating Brand Perception — Jaafar et al. (2024)
- AI-Optimized Customer Segmentation — Khair (2024)
- Scaling behaviour change for a 1.5 degree world — Newell et al. (2021)
- Towards the net zero carbon future: blockchain peer-to-peer carbon trading — Parhamfar et al. (2024)
- Mitigating Carbon Footprint through Green Software — Raghavasamy.M et al. (2024)
- Future Thinking Climate Change: Consumers Sustainability Perspectives — Rayburn et al. (2025)
- Prescriptive analytics for impulsive behaviour prevention — Raychaudhuri et al. (2021)
- Consumers behaviour towards carbon footprint labels on food — Rondoni & Grasso (2021)
- Multi-Product Category Choices Labeled for Ecological Footprints — Steiner et al. (2017)
- A Reference Architecture for Cloud–Edge Meta-Operating Systems — Trakadas et al. (2022)
- Eco-Wise Wallets: Personal Finance in the Age of Environmental Responsibility — WEN (2024)
- Research on energy consumption in household sector — Wang et al. (2024)
- Green behavior and perceived benefits in shaping green car buying behavior — Wu et al. (2024)
- Dual G.R.E.E.N Circle: generative AI for green consumption — Yin et al. (2025)
- Hospitality Industry 4.0 and Climate Change — Youssef & Zeqiri (2022)
- The Dual Role of AI in Developing Smart Cities — Zamponi & Barbierato (2022)
- Predicting Behaviour Patterns with AI Eye-Tracking — Ola et al. (2024)