★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★★Open research. Global perspectives.
The American Journal of Engineering and Technology

Volume 8.
Issue 11.

Volume 08 Issue 11

November 2026

Explore the research published in this issue. Read article details, abstracts and available full-text files.

Open access ISSN 2689-0984 4 articles
tajet ISSN 2689-0984

Ideas without
boundaries.

The American Journal of Engineering and Technology

VOLUME 8 / ISSUE 11 NOVEMBER 2026
IN THIS ISSUE

Table of contents.

4 articles

Engineering and Technology

4 articles
1
Engineering and Technology · OPEN ACCESS 01 November 2026

Conformal Gating Of Extracted Factor Vectors In LLM Signal Pipelines Under Delayed Labels And Regime Shift: A Simulation Study

Wrick Talukdar, Gowtham Shankar, Prasad Prabhu

Read Abstract

Language model pipelines extract multi-dimensional factor vectors from financial text, but a trading desk needs to know when an extracted vector can be trusted. Audit labels arrive late, errors cluster in stress regimes, and error scale shifts after model or prompt updates. We evaluate a conformal gate on the extraction layer in a seeded simulation (300 firms, 60 quarters, 200 replications) with heavy-tailed errors, regime stress, a structural break and a two-quarter audit lag. Per-dimension 90% intervals cover the full eight-dimension vector only 40 to 45% of the time, while a max-score construction reaches 89.7% before the break with intervals 18 to 26% narrower than Bonferroni. Calibration frozen at go-live falls to 78.2% joint coverage after the break. Rolling, weighted and delayed-feedback adaptive calibration recover 88.0 to 89.9% on average, but no label-driven method protects the quarters just after a shift. As a risk control, the adaptive gates reject most signals after the break, accepting only 9 to 14%, yet the error rate among accepted signals is still about 20%, against 28% without a gate, and gating lowers risk-adjusted return in this design. A label-free drift monitor flags 78.5% of abrupt breaks in the quarter they occur, against 3.0% for label-based monitoring, but only 3.5% under gradual drift, so its advantage is specific to abrupt breaks. To test whether these mechanisms appear with a real extractor, we run a pilot that extracts seven financial fields from the earnings press releases of 492 large US companies from an S&P 500 earnings-call dataset with an open-weight model and scores each figure against the SEC XBRL-reported value. Extraction is accurate for most figures, 96.3% within 1% once the reference labels are corrected and verified. Fixing our choice of revenue tag and checking every large-error label against the press release show that 38 of the 81 figures first scored as large errors rested on wrong labels rather than misreadings. The one self-reported signal collected, the model's GAAP flag, is constant and uninformative, while disagreement across three models detects 59% of the remaining large errors, and 74% of those confirmed as extraction errors, at the cost of flagging 8.6% of correct figures, which is consistent with the view that a usable confidence signal must come from a check outside a single model's output, although richer self-reported signals were not tested. A Monte Carlo resampling calibrated to the pilot, with the detector's operating point held fixed, gives sampling intervals for the accuracy and large-error rates. The results suggest three design rules: report joint reliability with a max-score rectangle, gate on the composite the operator consumes, and size positions by a smoothed scale rather than by calibrated width. We place the gate in a deployment framework and state its limits.

Read Article PDF
2
Engineering and Technology · OPEN ACCESS 12 November 2026

An Explainable AI Framework for Marketing Customer Segmentation and Campaign Response Prediction

Pritesh Gehlot, Soumyajit Mukherjee

Read Abstract

The segmentation of clients and the forecast of marketing campaign response are essential for optimising marketing expenditures and enhancing customer participation. This work suggests an XAI approach for segmenting the clients, predicting the responses using machine learning, and targeting them based on profit. Specifically, 2,055 cleansed customer records were extracted from the Kaggle Customer Personality Analysis dataset through preprocessing. We combine multiple clustering methods using consensus clustering to find the different segments of consumers, using Principal Component Analysis (PCA). The performance of seven machine learning classifiers is assessed using five-fold cross-validation, considering several performance metrics. The best accuracy (ACC) is 87.59% for Gradient Boosting (GB) and the ROC-AUC is 0.909 for Logistic Regression. The segment integration yields an improved recall (REC) rate of 79.4% and F1 score (F1) of 0.649, compared to the previous situation of REC value 73.0% and F1 value 0.622. The resulting profit-optimised solution is a projected profit of 61.1 currency units per 100 consumers. Moreover, elements such as catalogue purchases, recency, and campaign history are identified as elements that are significant via the techniques of SHapley Additive ExPlanations (SHAP) and permutation feature significance. The proposed approach will enable more accurate client segmentation, targeted marketing campaigns, and efficient use of resources. More testing is needed on a variety of real-life datasets to ensure greater generalisability, but the outcomes are encouraging.

Read Article PDF
3
Engineering and Technology · OPEN ACCESS 09 November 2026

Economic Efficiency of Introducing Digital Technologies in The Agriculture of Uzbekistan

Nuriddinova Ma'suda Lutfilla qizi

Read Abstract

Agriculture remains a fundamental pillar of the economy of Uzbekistan, providing employment for a significant portion of the population and contributing substantially to the gross domestic product (GDP). However, traditional agricultural practices in the region are increasingly challenged by critical water scarcity, land degradation, climate volatility, and fluctuating market demands. This research article comprehensively analyzes the economic efficiency of introducing advanced digital technologies—including Internet of Things (IoT) sensors, automated drip irrigation systems, Geographic Information Systems (GIS), drone-based remote sensing, and Big Data analytics—into Uzbekistan's agricultural matrix. Following the formal IMRAD framework, this study synthesizes empirical field data gathered over a multi-year period from select automated pilot farms in the Fergana and Namangan regions, comparing them against adjacent conventional farms. The empirical findings demonstrate a substantial increase in average crop yields (ranging from 18% to 25% for cotton and wheat), a profound reduction in water consumption (by 30% to 35%), and a noticeable decrease in chemical fertilizer and fuel deployment. Econometric modeling indicates an average Net Present Value (NPV) improvement and a typical payback period of 2.8 to 3.5 years for digital technology capital expenditures. Ultimately, this paper discusses the systemic institutional, technical, and educational barriers to widespread digital adoption, offering actionable policy recommendations for expanding precision agriculture across the Republic.

Read Article pdf
4
Engineering and Technology · OPEN ACCESS 10 November 2026

Methodology for Developing the Mathematical Competencies of Primary School Students with Hearing Impairments in Inclusive Education Through Visual-Didactic Technologies

Toshpulova Nodira, Qipchakova Xilola Qaddas qizi

Read Abstract

This article examines the pedagogical and methodological foundations for developing the mathematical competencies of primary school students with hearing impairments in an inclusive education environment. It analyzes the educational potential of visual-didactic technologies in the process of mastering mathematical knowledge, the criteria for selecting visual materials in accordance with students’ individual needs, and the methodological techniques that support the formation of mathematical concepts. A methodological approach based on the use of geometric models, digital images, color-coded diagrams, interactive tasks, and object-based practical activities is also proposed.

Read Article pdf