# Applied machine learning and data science: rigorous evaluation over headline accuracy

Research synthesis · Machine learning & data science
Canonical page: https://www.narasimharao.net/research/focus-areas/machine-learning-data-science/
Author: Narasimha Rao Vajjhala, Professor and Chair of Computer Science, American University in Bulgaria (ORCID 0000-0002-8260-2392)
Updated: 2026-10-04

> Vajjhala and colleagues show that in applied machine learning — from weed classification and weather prediction to medical imaging, software defects, project analytics and generative talking-face video — multi-seed statistical comparison, metrics beyond accuracy and leakage checks matter more than headline scores, and that simple models often match complex ones.

## Synthesis

Across applied studies in agriculture, meteorology, medical imaging, software engineering and business analytics, Narasimha Rao Vajjhala and colleagues have argued that the value of a machine learning model depends less on its headline accuracy than on how rigorously it is evaluated. [Kumar et al. (2026)](https://www.narasimharao.net/research/deep-learning-weed-classification-deepweeds-albania/) compared six deep learning configurations for nine-class weed classification on the DeepWeeds benchmark and demonstrated that, although fine-tuned DINOv2 and EfficientNet-B0 performed best, the differences among several high-performing models were not consistently statistically significant across random seeds. An exploratory test on Albanian field images exposed a clear geographic domain shift, showing that benchmark rankings do not transfer automatically to new farming regions.

A second recurring result is that simpler models are often sufficient. Using ten years of daily data for Sofia, Berlin and Tokyo, [Gigov et al. (2025)](https://www.narasimharao.net/research/hyperlocal-weather-temperature-humidity-prediction-machine-learning/) found that linear regression was statistically equivalent to random forest for temperature (R² > 0.994) at a fraction of the run time, and clearly better for humidity. The same pattern appears in project analytics: across 439 successful US government IT projects, linear regression predicted project manager commitment tenure better (r² = 0.238) than random forest or support vector machines ([Vajjhala & Strang, 2024a](https://www.narasimharao.net/research/machine-learning-project-manager-organizational-commitment-big-data/); [Strang & Vajjhala, 2025b](https://www.narasimharao.net/research/project-manager-commitment-machine-learning-big-data/)).

Vajjhala and colleagues also insist on metrics and diagnostics beyond accuracy. [Saheed et al. (2021)](https://www.narasimharao.net/research/ensemble-learning-software-defect-prediction/) evaluated a seven-model boosting and bagging ensemble for software defect prediction with AUC, F-measure and the Matthews correlation coefficient, reporting outstanding performance for ensemble CatBoost. [Haveri and Vajjhala (2026)](https://www.narasimharao.net/research/cnn-pneumonia-detection-chest-xray-efficientnetv2/) reported bootstrapped confidence intervals, calibration and CLAIM and TRIPOD-AI compliance for a pneumonia detector that reached an AUC of 0.967 and pneumonia recall of 0.98. Most pointedly, [Strang and Vajjhala (2026b)](https://www.narasimharao.net/research/regulatory-volatility-digital-supply-chains-esg-risk/) showed that a perfect in-sample classifier (AUC = 1.000) trained on supply chain records was an artefact of label leakage from near-duplicate items, and a companion study found that classifiers learned little from a firm’s own ESG ratings (kNN AUC = 0.497) ([Strang & Vajjhala, 2026a](https://www.narasimharao.net/research/machine-learning-esg-compliance-financial-software-projects/)). For generative models, [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/) reviewed 92 talking face generation studies (2016–2025) and found that evaluation still relies on PSNR, SSIM and SyncNet, none of which measures identity preservation, cross-cultural naturalness, trustworthiness or detectability.

Where the data are informative, the group has used machine learning to support decisions. Random forest analysis of about 17,430 US government IT projects identified seven failure indicators with 79.9% precision and a ROC area of 0.849 ([Strang & Vajjhala, 2023a](https://www.narasimharao.net/research/machine-learning-it-project-failure-indicators-big-data/)), and a random forest model best explained the performance gains of Albanian SMEs that piloted data analytics (R² = 0.661) ([Vajjhala & Strang, 2024b](https://www.narasimharao.net/research/data-analytics-sme-performance-albania/)). Earlier work applied correspondence analysis to 125,087 open global terrorism records ([Vajjhala et al., 2015](https://www.narasimharao.net/research/open-big-data-correspondence-analysis-global-terrorism/)), surveyed visual data mining for collaborative filtering ([Biba et al., 2017](https://www.narasimharao.net/research/visual-data-mining-collaborative-filtering-survey/)), and built recommender, web-prefetching and face-recognition systems ([Vajjhala et al., 2021b](https://www.narasimharao.net/research/twitter-profile-recommender-system-ibm-watson/); [Ajibesin et al., 2025](https://www.narasimharao.net/research/predictive-web-prefetching-clustering-weka-high-traffic/); [Imoh et al., 2023](https://www.narasimharao.net/research/deep-learning-face-recognition-missing-persons-nigeria/)).

Taken together, this body of work supports three practical claims for applied machine learning: compare models across repeated runs with statistical tests rather than single leaderboard scores; check training data for leakage before trusting near-perfect accuracy; and prefer the simpler, more interpretable model when performance is equivalent.

## Key claims with sources

1. Differences among high-performing deep learning models for weed classification were not consistently statistically significant across random seeds ([Kumar et al., 2026](https://www.narasimharao.net/research/deep-learning-weed-classification-deepweeds-albania/)).
2. Linear regression matched random forest for hyperlocal temperature prediction and outperformed it for humidity ([Gigov et al., 2025](https://www.narasimharao.net/research/hyperlocal-weather-temperature-humidity-prediction-machine-learning/)).
3. A perfect in-sample classifier (AUC = 1.000) on organizational records was an artefact of label leakage ([Strang & Vajjhala, 2026b](https://www.narasimharao.net/research/regulatory-volatility-digital-supply-chains-esg-risk/)).
4. A systematic review of 92 talking face generation studies found that the dominant evaluation metrics (PSNR, SSIM and SyncNet) measure none of identity preservation, cross-cultural naturalness, trustworthiness or detectability ([Salahudeen et al., 2026](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/)).

## The publications and what each contributes

| Publication | Method | Core finding |
|---|---|---|
| [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/) | Systematic literature review (Kitchenham and Charters; reported to PRISMA 2020) of 92 primary studies published 2016–2025, screened from 18,343 records in six repositories over four search iterations, with a five-criterion quality assessment and qualitative narrative synthesis across six research questions. | Across 92 primary studies (2016–2025), Diffusion Transformers and Gaussian Splatting account for about 36% of studies dated 2024 or later while GAN use has fallen below 3%; training data are demographically skewed and under-documented, and the dominant metrics (PSNR, SSIM, SyncNet) do not measure what deployment requires. |
| [Strang and Vajjhala (2026b)](https://www.narasimharao.net/research/regulatory-volatility-digital-supply-chains-esg-risk/) | Quantitative retrospective analysis of archival organisational records (N = 1,988): descriptive statistics, exploratory factor analysis with parallel analysis, and binary logistic regression with classification diagnostics. | Perceived regulatory-volatility risks do not form a single construct (KMO = 0.470), and labor and environmental risks were rated lowest of all (medians 0 and 1 on 0–5 scales) during a period of record forced-labor enforcement. |
| [Strang and Vajjhala (2026a)](https://www.narasimharao.net/research/machine-learning-esg-compliance-financial-software-projects/) | Single-case socio-technical systems study of archival project records (N = 207) using an open-source data science workflow: correlation analysis, principal component analysis, common-method-bias diagnostics, and kNN and SVM classification. | In 207 financial software project records, stakeholder-rated social and governance factors tracked the overall project score almost perfectly (r = +0.995 and +0.966) while the environmental factor was unrelated to it, and machine learning classifiers performed weakly (kNN AUC = 0.497; SVM accuracy 61.8%). |
| [Kumar et al. (2026)](https://www.narasimharao.net/research/deep-learning-weed-classification-deepweeds-albania/) | Comparative deep learning experiment on the DeepWeeds benchmark (nine classes) with leakage-checked splits, repeated multi-seed training of six architectures, statistical model comparison, and an exploratory out-of-distribution probe on unlabelled Albanian field images. | Fine-tuned DINOv2 and EfficientNet-B0 performed best on nine-class DeepWeeds data, but differences among several high-performing models were not consistently statistically significant across random seeds. |
| [Haveri and Vajjhala (2026)](https://www.narasimharao.net/research/cnn-pneumonia-detection-chest-xray-efficientnetv2/) | Binary medical image classification with an EfficientNetV2B0 convolutional neural network and ImageNet transfer learning on the Kermany paediatric chest X-ray dataset, following CLAIM and TRIPOD-AI reporting, with bootstrapped confidence intervals, DeLong test, calibration metrics and Grad-CAM. | A lightweight EfficientNetV2B0 model trained under CLAIM and TRIPOD-AI reporting standards achieved an AUC of 0.967 (95% CI 0.953–0.979) and pneumonia recall of 0.98 on paediatric chest X-rays, missing only about 2% of true cases. |
| [Gigov et al. (2025)](https://www.narasimharao.net/research/hyperlocal-weather-temperature-humidity-prediction-machine-learning/) | Comparative supervised regression experiment (linear, polynomial, decision tree, random forest) on ten years of daily Open-Meteo data for Sofia, Berlin and Tokyo with a chronological 7:3 split, grid-search tuning, and R², MAE, RMSE and execution-time evaluation. | On ten years of daily data for Sofia, Berlin and Tokyo, linear regression matched random forest for temperature at a fraction of the run time and clearly outperformed it for humidity (R² 0.940–0.964), while decision trees performed worst. |
| [Strang and Vajjhala (2025b)](https://www.narasimharao.net/research/project-manager-commitment-machine-learning-big-data/) | Exploratory machine learning study of retrospective secondary big data (439 successful US government IT projects, 1,230 features): multidimensional scaling, silhouette scores, and linear regression, random forest and SVM regression with ten-fold cross-validation. | Across 439 successful US government IT projects with 1,230 features, linear regression outperformed random forest and SVM in predicting project manager commitment tenure. |
| [Ajibesin et al. (2025)](https://www.narasimharao.net/research/predictive-web-prefetching-clustering-weka-high-traffic/) | Design and evaluation of a web prefetching model: preprocessing of server log files, construction of an online navigation graph, clustering with the WEKA machine learning toolkit, and comparison against conventional prefetching techniques on domain client-group logs. | A prefetching approach that builds an online navigation graph from server logs and clusters navigation behaviour with WEKA outperformed conventional prefetching techniques on domain client-group logs, particularly once server idle time allowed prefetching across all user predictions. |
| [Vajjhala and Strang (2024a)](https://www.narasimharao.net/research/machine-learning-project-manager-organizational-commitment-big-data/) | Exploratory pragmatic machine learning study of a declassified U.S. government IT project repository (439 successful projects, 1,230 features) using radial analysis, multidimensional scaling and supervised regression (linear, random forest, SVM) with relief-based feature ranking. | Supervised machine learning on 439 successful US government IT projects showed linear regression best predicted project manager commitment tenure (r² = 0.238), with line of business, experience, network collaboration tools and cross-industry team membership as the strongest features — supporting an association between higher commitment and project success. |
| [Vajjhala and Strang (2024b)](https://www.narasimharao.net/research/data-analytics-sme-performance-albania/) | Quantitative post-positivist survey of 41 Albanian SMEs: descriptive statistics, Pearson and Spearman correlation, generalised linear model, MANOVA, and random forest, SVM, linear regression and neural network models. | Among 41 Albanian SMEs, data-analytics use correlated positively with operational efficiency, process optimisation, profitability and market growth (r = 0.540 to 0.769), with a significant multivariate effect (MANOVA F(1, 38) = 20.029, p < 0.001); a random forest model explained the outcomes best (R² = 0.661). |
| [Strang and Vajjhala (2023a)](https://www.narasimharao.net/research/machine-learning-it-project-failure-indicators-big-data/) | Pragmatic recursive mixed-methods design on U.S. Government Accountability Office big data (about 17,430 IT projects): structured data cleaning, random forest classification of project breach/failure, and post-hoc logistic regression in SPSS. | Random forest analysis of about 17,430 US government IT projects identified seven failure indicators with 81% recall and a ROC area of 0.849. |
| [Imoh et al. (2023)](https://www.narasimharao.net/research/deep-learning-face-recognition-missing-persons-nigeria/) | Experimental system design: face recognition with a convolutional neural network, facial calibration and modelling for feature extraction, and comparison of face encodings against a query image. | The authors built an experimental face-recognition system that combines a convolutional neural network with facial calibration and modelling to extract face encodings and match them against a query image, as a faster tool for identifying missing persons. |
| [Saheed et al. (2021)](https://www.narasimharao.net/research/ensemble-learning-software-defect-prediction/) | Experimental machine learning comparison: seven ensemble models (CatBoost, LightGBM, XGBoost and boosted/bagged variants) plus a logistic regression baseline evaluated on six software defect datasets (NASA repository) with AUC, precision, recall, F-measure and Matthews correlation coefficient. | A seven-model ensemble of boosted and bagged learners, evaluated with AUC, precision, recall, F-measure and Matthews correlation coefficient on NASA defect datasets, outperformed a logistic regression baseline, and the ensemble CatBoost model gave outstanding performance on all three datasets reported. |
| [Vajjhala et al. (2021b)](https://www.narasimharao.net/research/twitter-profile-recommender-system-ibm-watson/) | System design and experimental evaluation: mining and analysis of Twitter profiles and timelines on the IBM Watson platform to predict product and service categories, with a reported correlation between consumed categories and tweet content. | A recommender built on IBM Watson that mines a user’s Twitter profile and timeline predicted the category of goods and services the user is most likely to consume, and the authors report a strong correlation between consumed categories and tweet content. |
| [Biba et al. (2017)](https://www.narasimharao.net/research/visual-data-mining-collaborative-filtering-survey/) | State-of-the-art literature survey of visual data mining taxonomies and tools, memory-based, model-based and hybrid collaborative filtering, and visual collaborative filtering systems evaluated on EachMovie, MovieLens and Netflix data. | The survey organizes visual data mining techniques, tools and collaborative filtering approaches, and concludes that combining visual data mining with collaborative filtering has led to substantial improvement and can alleviate the shortcomings of traditional collaborative filtering. |
| [Vajjhala et al. (2015)](https://www.narasimharao.net/research/open-big-data-correspondence-analysis-global-terrorism/) | Positivist exploratory case study applying simple correspondence analysis in SPSS 21 to an open big dataset of 125,087 global terrorism records (1970–2013), relating nine attack methods to 13 world regions. | Simple correspondence analysis in standard statistical software, applied to 125,087 global terrorism records (1970–2013), related nine attack methods to 13 world regions with two dimensions capturing 94.5% of inertia — demonstrating an accessible method for visualizing open big data. |

## Key terms in this area

- **Out-of-distribution (OOD) data:** Data that differ systematically from a model’s training data — here, field images from a different country, soil, and lighting than the benchmark. (as used in [Kumar et al. (2026)](https://www.narasimharao.net/research/deep-learning-weed-classification-deepweeds-albania/))
- **Multi-seed evaluation:** Training and testing each model several times with different random initializations to check whether performance differences are stable rather than due to chance. (as used in [Kumar et al. (2026)](https://www.narasimharao.net/research/deep-learning-weed-classification-deepweeds-albania/))
- **Linear probing vs. fine-tuning:** Linear probing trains only a new output layer on top of a frozen pretrained model; full fine-tuning updates all of the model’s weights. (as used in [Kumar et al. (2026)](https://www.narasimharao.net/research/deep-learning-weed-classification-deepweeds-albania/))
- **Hyperlocal weather prediction:** Forecasting weather conditions at a fine spatial scale such as a city or community, rather than the coarse grids (e.g. 15 km × 15 km) used by global forecasting systems. (as used in [Gigov et al. (2025)](https://www.narasimharao.net/research/hyperlocal-weather-temperature-humidity-prediction-machine-learning/))
- **Random forest regression:** An ensemble of decision trees trained on bootstrapped samples and random feature subsets whose predictions are averaged to give the final prediction. (as used in [Gigov et al. (2025)](https://www.narasimharao.net/research/hyperlocal-weather-temperature-humidity-prediction-machine-learning/))
- **Coefficient of determination (R²):** A measure of how much variance in the target is explained by the model, where values close to 1 indicate a near-perfect fit. (as used in [Gigov et al. (2025)](https://www.narasimharao.net/research/hyperlocal-weather-temperature-humidity-prediction-machine-learning/))
- **Regulatory volatility:** Rapid, hard-to-predict change in the rules that govern trade and operations — tariffs, customs regimes, labor and environmental (ESG) regulation — that makes supply chains less predictable. (as used in [Strang and Vajjhala (2026b)](https://www.narasimharao.net/research/regulatory-volatility-digital-supply-chains-esg-risk/))
- **Sustainability leakage:** As used in this study: when surprise ESG enforcement leads firms to exit suppliers or move sourcing elsewhere instead of remediating conditions in the scrutinized region. (as used in [Strang and Vajjhala (2026b)](https://www.narasimharao.net/research/regulatory-volatility-digital-supply-chains-esg-risk/))
- **Label leakage:** When a predictor variable effectively encodes the outcome being predicted, producing unrealistically perfect model accuracy that will not hold in real use. (as used in [Strang and Vajjhala (2026b)](https://www.narasimharao.net/research/regulatory-volatility-digital-supply-chains-esg-risk/))
- **Talking face generation (TFG):** The synthesis of photorealistic, temporally coherent video in which a face speaks in synchrony with a given audio signal, typically from a single portrait image and an audio clip. (as used in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/))
- **Diffusion Transformer (DiT):** A diffusion model in which the usual U-Net denoising network is replaced by a transformer operating over latent tokens, so that output quality scales with model size and compute. (as used in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/))
- **Gaussian Splatting:** A 3D rendering approach that represents a head or scene as learned 3D Gaussians and draws them by fast differentiable rasterisation instead of volume rendering, which makes real-time talking-head synthesis possible. (as used in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/))
- **Content provenance (C2PA):** A cryptographically signed manifest, specified by the Coalition for Content Provenance and Authenticity (C2PA v2.0), recording how a media asset was produced and bound to that asset, so that the authenticity and integrity of the claim can be verified. (as used in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/))
- **Deepfake Detectability Rate (DDR):** A metric proposed in this review: the proportion of generated frames correctly classified as synthetic by an ensemble of state-of-the-art deepfake detectors, intended to operationalise the machine-detectability required by EU AI Act Article 50(2). (as used in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/))
- **Systemic embeddedness:** The sixth principle the RTFG-STS Framework adds to the five AI4People principles: the design and governance of a generative system should reference the concrete socio-technical context in which it is deployed, not the model in isolation. (as used in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/))

## Standards, datasets and methods named in these studies

- [DeepWeeds](https://doi.org/10.1038/s41598-018-38343-3) (Benchmark dataset). Olsen, A., Konovalov, D. A., Philippa, B., et al. (2019). DeepWeeds: A multiclass weed species image dataset for deep learning. Scientific Reports, 9, 2058. https://doi.org/10.1038/s41598-018-38343-3 Named in [Kumar et al. (2026)](https://www.narasimharao.net/research/deep-learning-weed-classification-deepweeds-albania/).
- [DINOv2](https://arxiv.org/abs/2304.07193) (Model). Oquab, M., Darcet, T., Moutakanni, T., et al. (2023). DINOv2: Learning robust visual features without supervision. arXiv:2304.07193. Named in [Kumar et al. (2026)](https://www.narasimharao.net/research/deep-learning-weed-classification-deepweeds-albania/).
- [EfficientNet](https://proceedings.mlr.press/v97/tan19a.html) (Model). Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking model scaling for convolutional neural networks. Proceedings of the 36th International Conference on Machine Learning (PMLR 97). Named in [Kumar et al. (2026)](https://www.narasimharao.net/research/deep-learning-weed-classification-deepweeds-albania/).
- [Open-Meteo Historical Weather API](https://open-meteo.com/en/docs/historical-weather-api) (Data source). Open-Meteo. Historical Weather API. Named in [Gigov et al. (2025)](https://www.narasimharao.net/research/hyperlocal-weather-temperature-humidity-prediction-machine-learning/).
- [CatBoost](https://proceedings.neurips.cc/paper/2018/hash/14491b756b3a51daac41c24863285549-Abstract.html) (Method). Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A. V., & Gulin, A. (2018). CatBoost: Unbiased boosting with categorical features. Advances in Neural Information Processing Systems 31. Named in [Saheed et al. (2021)](https://www.narasimharao.net/research/ensemble-learning-software-defect-prediction/).
- [LightGBM](https://proceedings.neurips.cc/paper/2017/hash/6449f44a102fde848669bdd9eb6b76fa-Abstract.html) (Method). Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., & Liu, T.-Y. (2017). LightGBM: A highly efficient gradient boosting decision tree. Advances in Neural Information Processing Systems 30. Named in [Saheed et al. (2021)](https://www.narasimharao.net/research/ensemble-learning-software-defect-prediction/).
- [XGBoost](https://doi.org/10.1145/2939672.2939785) (Method). Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794. https://doi.org/10.1145/2939672.2939785 Named in [Saheed et al. (2021)](https://www.narasimharao.net/research/ensemble-learning-software-defect-prediction/).
- [Kermany paediatric chest X-ray dataset](https://doi.org/10.1016/j.cell.2018.02.010) (Dataset). Kermany, D. S., Goldbaum, M., Cai, W., Valentim, C. C. S., Liang, H., Baxter, S. L., et al. (2018). Identifying medical diagnoses and treatable diseases by image-based deep learning. Cell, 172(5), 1122–1131.e9. https://doi.org/10.1016/j.cell.2018.02.010 Named in [Haveri and Vajjhala (2026)](https://www.narasimharao.net/research/cnn-pneumonia-detection-chest-xray-efficientnetv2/).
- [EfficientNetV2](https://proceedings.mlr.press/v139/tan21a.html) (Model). Tan, M., & Le, Q. V. (2021). EfficientNetV2: Smaller models and faster training. Proceedings of the 38th International Conference on Machine Learning (PMLR 139). Named in [Haveri and Vajjhala (2026)](https://www.narasimharao.net/research/cnn-pneumonia-detection-chest-xray-efficientnetv2/).
- [ImageNet](https://doi.org/10.1109/CVPR.2009.5206848) (Dataset). Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009). ImageNet: A large-scale hierarchical image database. 2009 IEEE Conference on Computer Vision and Pattern Recognition, 248–255. https://doi.org/10.1109/CVPR.2009.5206848 Named in [Haveri and Vajjhala (2026)](https://www.narasimharao.net/research/cnn-pneumonia-detection-chest-xray-efficientnetv2/).
- [CLAIM — Checklist for Artificial Intelligence in Medical Imaging](https://doi.org/10.1148/ryai.2020200029) (Reporting guideline). Mongan, J., Moy, L., & Kahn, C. E., Jr. (2020). Checklist for Artificial Intelligence in Medical Imaging (CLAIM): A guide for authors and reviewers. Radiology: Artificial Intelligence, 2(2), e200029. https://doi.org/10.1148/ryai.2020200029 Named in [Haveri and Vajjhala (2026)](https://www.narasimharao.net/research/cnn-pneumonia-detection-chest-xray-efficientnetv2/).
- [TRIPOD+AI statement](https://pmc.ncbi.nlm.nih.gov/articles/PMC11019967/) (Reporting guideline). Collins, G. S., Moons, K. G. M., Dhiman, P., Riley, R. D., Beam, A. L., Van Calster, B., et al. (2024). TRIPOD+AI statement: Updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ, 385, e078378. https://doi.org/10.1136/bmj-2023-078378 Named in [Haveri and Vajjhala (2026)](https://www.narasimharao.net/research/cnn-pneumonia-detection-chest-xray-efficientnetv2/).
- [Guidelines for performing systematic literature reviews in software engineering](https://ebse.webspace.durham.ac.uk/ebse-bibliography/guidelines-for-performing-systematic-literature-reviews-in-software-engineering/) (Review guideline). Kitchenham, B., & Charters, S. (2007). Guidelines for performing systematic literature reviews in software engineering (EBSE Technical Report EBSE-2007-01). Keele University and Durham University. Named in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/).
- [PRISMA 2020 statement](https://doi.org/10.1136/bmj.n71) (Reporting guideline). Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., et al. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, n71. https://doi.org/10.1136/bmj.n71 Named in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/).
- [EU Artificial Intelligence Act — Regulation (EU) 2024/1689](https://eur-lex.europa.eu/eli/reg/2024/1689/oj/eng) (Regulation). Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act). Official Journal of the European Union. Named in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/).
- [C2PA Technical Specification (Content Credentials)](https://spec.c2pa.org/specifications/specifications/2.4/specs/C2PA_Specification.html) (Technical standard). Coalition for Content Provenance and Authenticity (C2PA). Content Credentials: C2PA Technical Specification. Named in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/).
- [TAKE IT DOWN Act (United States, Public Law 119-12)](https://www.congress.gov/bill/119th-congress/senate-bill/146) (Law). TAKE IT DOWN Act, S. 146, 119th Congress, Public Law 119-12 (2025). Named in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/).
- [AI4People ethical framework](https://doi.org/10.1007/s11023-018-9482-5) (Ethics framework). Floridi, L., Cowls, J., Beltrametti, M., Chatila, R., Chazerand, P., Dignum, V., et al. (2018). AI4People—An ethical framework for a good AI society: Opportunities, risks, principles, and recommendations. Minds and Machines, 28, 689–707. https://doi.org/10.1007/s11023-018-9482-5 Named in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/).
- [VoxCeleb2](https://www.isca-archive.org/interspeech_2018/chung18b_interspeech.html) (Dataset). Chung, J. S., Nagrani, A., & Zisserman, A. (2018). VoxCeleb2: Deep speaker recognition. Proceedings of Interspeech 2018, 1086–1090. Named in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/).
- [CREMA-D](https://doi.org/10.1109/TAFFC.2014.2336244) (Dataset). Cao, H., Cooper, D. G., Keutmann, M. K., Gur, R. C., Nenkova, A., & Verma, R. (2014). CREMA-D: Crowd-sourced emotional multimodal actors dataset. IEEE Transactions on Affective Computing, 5(4), 377–390. https://doi.org/10.1109/TAFFC.2014.2336244 Named in [Salahudeen et al. (2026)](https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/).
- [WEKA machine learning workbench](https://ml.cms.waikato.ac.nz/weka/) (Software). University of Waikato. Weka 3: Machine learning software in Java. Named in [Ajibesin et al. (2025)](https://www.narasimharao.net/research/predictive-web-prefetching-clustering-weka-high-traffic/).

## References

- Ajibesin, A. A., Vajjhala, N. R., Joel, E., & Rakshit, S. (2025). Predictive Web Prefetching: A Combined Approach Using Clustering Algorithms and WEKA in High-Traffic Settings. In F. Lin, D. Pastor, N. Kesswani, A. Patel, S. Bordoloi, & C. Koley (Eds.), Artificial Intelligence in Internet of Things (IoT): Key Digital Trends: Proceedings of 8th International Conference on Internet of Things and Connected Technologies (ICIoTCT 2023) (pp. 221–231). Springer Nature Singapore. https://doi.org/10.1007/978-981-97-5786-2_17 — summary and key findings: https://www.narasimharao.net/research/predictive-web-prefetching-clustering-weka-high-traffic/
- Biba, M., Vajjhala, N. R., & Nishani, L. (2017). Visual Data Mining for Collaborative Filtering: A State-of-the-Art Survey. In Collaborative Filtering Using Data Mining and Analysis (pp. 217–235). IGI Global. https://doi.org/10.4018/978-1-5225-0489-4.ch012 — summary and key findings: https://www.narasimharao.net/research/visual-data-mining-collaborative-filtering-survey/
- Gigov, L., Vajjhala, N. R., & Stoilov, A. (2025). Hyperlocal Temperature and Humidity Prediction Using Supervised Machine Learning. In 2025 2nd Global AI Summit – International Conference on Artificial Intelligence and Emerging Technology (AI Summit) (pp. 940–945). IEEE. https://doi.org/10.1109/AISummit66170.2025.11411099 — summary and key findings: https://www.narasimharao.net/research/hyperlocal-weather-temperature-humidity-prediction-machine-learning/
- Haveri, K., & Vajjhala, N. R. (2026). Deep Learning for Medical Image Analysis: CNN-based Pneumonia Detection on Chest X-Rays. In 2026 6th International Conference on Pervasive Computing and Social Networking (ICPCSN) (pp. 156–161). IEEE. https://doi.org/10.1109/ICPCSN68523.2026.11543623 — summary and key findings: https://www.narasimharao.net/research/cnn-pneumonia-detection-chest-xray-efficientnetv2/
- Imoh, N., Vajjhala, N. R., & Rakshit, S. (2023). Experimental Face Recognition Using Applied Deep Learning Approaches to Find Missing Persons. In S. Basu, D. K. Kole, A. K. Maji, D. Plewczynski, & D. Bhattacharjee (Eds.), Proceedings of International Conference on Frontiers in Computing and Systems: COMSYS 2021 (pp. 3–11). Springer Nature Singapore. https://doi.org/10.1007/978-981-19-0105-8_1 — summary and key findings: https://www.narasimharao.net/research/deep-learning-face-recognition-missing-persons-nigeria/
- Kumar, R., Vajjhala, N. R., Ramollari, E., & Joca, E. (2026). A Comparative Evaluation of Deep Learning Architectures for Weed Classification, with an Exploratory Out-of-Distribution Analysis of Albanian Field Images. Information, 17(8), Article 751. https://doi.org/10.3390/info17080751 — summary and key findings: https://www.narasimharao.net/research/deep-learning-weed-classification-deepweeds-albania/
- Saheed, Y. K., Longe, O., Baba, U. A., Rakshit, S., & Vajjhala, N. R. (2021). An Ensemble Learning Approach for Software Defect Prediction in Developing Quality Software Product. In M. Singh, V. Tyagi, P. K. Gupta, J. Flusser, T. Ören, & V. R. Sonawane (Eds.), Advances in Computing and Data Sciences: 5th International Conference, ICACDS 2021, Nashik, India, April 23–24, 2021, Revised Selected Papers, Part I (pp. 317–326). Springer International Publishing. https://doi.org/10.1007/978-3-030-81462-5_29 — summary and key findings: https://www.narasimharao.net/research/ensemble-learning-software-defect-prediction/
- Salahudeen, R., Fonkam, M., Vajjhala, N. R., Junaidu, S. B., & Garba, A. (2026). Talking Face Generation in Socio-Technical Systems: A Systematic Review of Deep Learning Architectures, Deployment Contexts, and Governance Frameworks. Systems, 14(10), Article 1198. https://doi.org/10.3390/systems14101198 — summary and key findings: https://www.narasimharao.net/research/talking-face-generation-socio-technical-systems-systematic-review/
- Strang, K. D., & Vajjhala, N. R. (2023a). Mining Project Failure Indicators From Big Data Using Machine Learning Mixed Methods. International Journal of Information Technology Project Management, 14(1), 1–24. https://doi.org/10.4018/IJITPM.317221 — summary and key findings: https://www.narasimharao.net/research/machine-learning-it-project-failure-indicators-big-data/
- Strang, K. D., & Vajjhala, N. R. (2025b). Exploring project manager commitment using machine learning on fuzzy big data. International Journal of Project Organisation and Management, 17(2), 135–152. https://doi.org/10.1504/IJPOM.2025.146727 — summary and key findings: https://www.narasimharao.net/research/project-manager-commitment-machine-learning-big-data/
- Strang, K. D., & Vajjhala, N. R. (2026a). Machine Learning and Data Science for ESG Compliance Measurement in Financial Software Engineering Projects: A Single-Case Socio-Technical Systems Analysis. Systems, 14(8), Article 990. https://doi.org/10.3390/systems14080990 — summary and key findings: https://www.narasimharao.net/research/machine-learning-esg-compliance-financial-software-projects/
- Strang, K. D., & Vajjhala, N. R. (2026b). Regulatory Volatility in Digital Supply Chains: An Information Systems Analytics Study of Decision-Maker Risk Perceptions and Sustainability-Related Outcomes. Sustainability, 18(18), Article 9359. https://doi.org/10.3390/su18189359 — summary and key findings: https://www.narasimharao.net/research/regulatory-volatility-digital-supply-chains-esg-risk/
- Vajjhala, N. R., & Strang, K. D. (2024a). An Exploratory Big Data Approach to Understanding Commitment in Projects. In Á. Rocha (Ed.), WorldCIST 2024 (World Conference on Information Systems and Technologies) (pp. 66–75). Springer Nature Switzerland AG. https://doi.org/10.1007/978-3-031-60227-6_6 — summary and key findings: https://www.narasimharao.net/research/machine-learning-project-manager-organizational-commitment-big-data/
- Vajjhala, N. R., & Strang, K. D. (2024b). Profitability, effectiveness, operational efficiency, and market growth of SMEs in Albania after piloting data analytics. International Journal of Services and Standards, 14(1), 51–64. https://doi.org/10.1504/IJSS.2024.140078 — summary and key findings: https://www.narasimharao.net/research/data-analytics-sme-performance-albania/
- Vajjhala, N. R., Rakshit, S., Oshogbunu, M., & Salisu, S. (2021b). Novel User Preference Recommender System Based on Twitter Profile Analysis. In S. Borah, R. Pradhan, N. Dey, & P. Gupta (Eds.), Soft Computing Techniques and Applications: Proceeding of the International Conference on Computing and Communication (IC3 2020) (pp. 85–93). Springer Singapore. https://doi.org/10.1007/978-981-15-7394-1_7 — summary and key findings: https://www.narasimharao.net/research/twitter-profile-recommender-system-ibm-watson/
- Vajjhala, N. R., Strang, K. D., & Sun, Z. (2015). Statistical Modeling and Visualizing Open Big Data Using a Terrorism Case Study. In 2015 3rd International Conference on Future Internet of Things and Cloud (FiCloud), Rome, Italy (pp. 489–496). IEEE. https://doi.org/10.1109/FiCloud.2015.15 — summary and key findings: https://www.narasimharao.net/research/open-big-data-correspondence-analysis-global-terrorism/