(Created page with " == Abstract == <p>This study presents a hybrid automated framework based on a combination of machine learning (ML) and natural language processing (NLP) approaches for the a...")
 
 
(One intermediate revision by the same user not shown)
Line 2: Line 2:
 
== Abstract ==
 
== Abstract ==
  
−
<p>This study presents a hybrid automated framework based on a combination of machine learning (ML) and natural language processing (NLP) approaches for the automatic categorization and extraction of nonfunctional requirements (NFRs) from free-text software development documents. Using the PROMISE dataset, this framework systematically integrates semantic representation learning, deep feature extraction, and kernel-based classification to improve the performance of NFR classification. Unlike current CNN-based approaches with end-to-end softmaxbased classification, our proposed method fundamentally decouples feature learning from decision making. The first approach is to use Word2Vec embeddings to capture semantic context, and then use Convolutional Neural Networks (CNNs) as high-level feature extractors. An Improved Support Vector Machine with a Radial Basis Function kernel (ISVM-RBF) is applied for final classification, enabling more discriminative decision boundaries to be drawn in the high-dimensional semantic feature space. We reveal a considerable performance improvement with the CNN&ndash; Word2Vec setup, achieving as high as a 90% precision, significantly outperforming standard ML classifiers. The study points to three main findings: (i) CNN-based feature extraction is an efficient approach for finding and classifying NFRs, (ii) the semantic representation provided by word embedding methods is clearly superior to other traditional methods used in NLP, and (iii) NLP preprocessing of text is crucial for enhancing classification accuracy. Finally, ISVM-RBF adapts kernel-based classification over features derived from CNN, which enhances the robustness of the model to semantic overlaps between NFR categories and alleviates challenges posed by potentially large textual datasets required to train such models. This hybrid CNN&ndash;ISVM-RBF design constitutes the methodological novelty of the proposed method and effectively distinguishes it from current state-of-the-art methods in the literature.OPEN ACCESS Received: 03/11/2025 Accepted: 22/01/2026 Published: 16/04/2026</p>
+
<p>In managing strong roof loading in steep-inclined longwall panels, this
−
 
+
study adopts partial gob backfill mining along the dip direction. Four
−
 
+
controlling factors for roof deformation are identified: working face
 +
length (L), mining depth (H), seam dip angle (α), and backfill length
 +
(a). Parametric analysis determines that L = 105 m combined with a
 +
2/5 backfill ratio achieves optimal strata control. Physical experiments
 +
recorded dip-direction stress gradients: upper (8.65/7.79/8.45 MPa), central
 +
peak (9.86/9.15/9.86 MPa), and lower (8.82/8.41/8.83 MPa), with displacement
 +
increments of horizontal (+115.6%/+73.9%/+74.1%), vertical
 +
(+136.2%/+48.9%/+21.3%), and resultant (+80.6%/+94.8%/+39.2%).
 +
FLAC3D simulations systematically varied backfill ratios (1/5, 2/5, 3/5)
 +
and face lengths (90, 105, 120 m). Increasing the ratio from 1/5 to 2/5
 +
reduced peak stress by 7.7% (15.65 → 14.45 MPa) and subsidence by
 +
39.3% (1.78 → 1.08 m), while further increase to 3/5 yielded marginal
 +
gains (4.5%, 31.5%). At the optimal 2/5 ratio, extending face length from
 +
90 to 105 m increased abutment stress by 8.9% (13.27→14.45 MPa) and
 +
subsidence by 17.4% (0.92→1.08 m), while 120mcaused disproportionate
 +
surges (5.2%, 49.1%) with plastic zone height soaring 81.9% (36.05→
 +
65.56 m). Under the optimal 105 m–2/5 configuration, staged advance
 +
(20–80 m) quantified progressive stress transfer: lower-end pillar stress
 +
rose 20.4% (9.22→11.10MPa), backfill stress 24.7% (8.75→10.91MPa),
 +
and roof subsidence from 302 to 688 mm, with plastic zone evolving as
 +
an asymmetric arch characterized by shear failure at the arch foot (lower
 +
pillar/backfill interface) and tensile failure at the crown. This integrated
 +
approach confirms that partial backfill effectively regulates strata behavior,
 +
providing a quantitative framework for sustainable steep-seam mining.</p>
  
 
== Document ==
 
== Document ==
−
<pdf>Media:Draft_content_853872237-9046-document.pdf</pdf>
+
<pdf>Media:Review_374494903953_4281_89. TSP_RIMNI_77102.pdf</pdf>

Latest revision as of 10:24, 22 May 2026

Abstract

In managing strong roof loading in steep-inclined longwall panels, this study adopts partial gob backfill mining along the dip direction. Four controlling factors for roof deformation are identified: working face length (L), mining depth (H), seam dip angle (α), and backfill length (a). Parametric analysis determines that L = 105 m combined with a 2/5 backfill ratio achieves optimal strata control. Physical experiments recorded dip-direction stress gradients: upper (8.65/7.79/8.45 MPa), central peak (9.86/9.15/9.86 MPa), and lower (8.82/8.41/8.83 MPa), with displacement increments of horizontal (+115.6%/+73.9%/+74.1%), vertical (+136.2%/+48.9%/+21.3%), and resultant (+80.6%/+94.8%/+39.2%). FLAC3D simulations systematically varied backfill ratios (1/5, 2/5, 3/5) and face lengths (90, 105, 120 m). Increasing the ratio from 1/5 to 2/5 reduced peak stress by 7.7% (15.65 → 14.45 MPa) and subsidence by 39.3% (1.78 → 1.08 m), while further increase to 3/5 yielded marginal gains (4.5%, 31.5%). At the optimal 2/5 ratio, extending face length from 90 to 105 m increased abutment stress by 8.9% (13.27→14.45 MPa) and subsidence by 17.4% (0.92→1.08 m), while 120mcaused disproportionate surges (5.2%, 49.1%) with plastic zone height soaring 81.9% (36.05→ 65.56 m). Under the optimal 105 m–2/5 configuration, staged advance (20–80 m) quantified progressive stress transfer: lower-end pillar stress rose 20.4% (9.22→11.10MPa), backfill stress 24.7% (8.75→10.91MPa), and roof subsidence from 302 to 688 mm, with plastic zone evolving as an asymmetric arch characterized by shear failure at the arch foot (lower pillar/backfill interface) and tensile failure at the crown. This integrated approach confirms that partial backfill effectively regulates strata behavior, providing a quantitative framework for sustainable steep-seam mining.

Document

The PDF file did not load properly or your web browser does not support viewing PDF files. Download directly to your device: Download PDF document
Back to Top
GET PDF

Document information

Published on 16/04/26

Licence: CC BY-NC-SA license

Document Score

0

Views 0
Recommendations 0

Share this document

Keywords

• • • • •

claim authorship

Are you one of the authors of this document?

I am an author