Metin madenciliği ve makine öğrenimi tekniklerinin uygulanmasıyla standart inşaat sözleşmesi belgelerinin metin analizi
Küçük Resim Yok
Tarih
2025
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Düzce Üniversitesi
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Bu tez çalışmasında, standart Tasarla-İnşa Et (Design-Build) sözleşmelerinin çok boyutlu içerik yapısını anlamlandırmaya yönelik olarak doğal dil işleme (NLP), metin madenciliği (text mining) ve makine öğrenimi (ML) tekniklerine dayalı bütüncül bir sınıflandırma çerçevesi geliştirilmiştir. Tasarla - İnşa Et proje teslim yöntemi özelinde, 'FIDIC Conditions of Contract for Design, Build and Operate' ve 'JCT Design and Build Conract' standart İngilizce sözleşme metinleri üzerinde yürütülen çalışmada, sözleşme maddeleri hem tematik (yükümlülükler, yapılacaklar, isteğe bağlı ifadeler), hem de işlevsel (maliyet, süre, kalite) eksenlerde sınıflandırılmıştır. Çalışma üç aşamalı deneysel yapı üzerine inşa edilmiştir. İlk aşamada, sözleşme hükümleri "obligations", "to do", "optionals" ve "other general statements" gibi tematik sınıflara ayrılarak, klasik metin temsil yöntemleri (TF-IDF, BoW, Word2Vec: CBOW & Skip-Gram) ile birlikte "Destek Vektör Makineleri (SVM), Lojistik Regresyon (LR), Saf Bayes (NB), K-en yakın komşular (KNN), Karar Ağaçları (DT)" gibi temel Makine Öğrenimi (ML) algoritmaları ve "XGBoost, AdaBoost, Hard Voting, Soft Voting, Stacking" gibi Topluluk Öğrenmesi (EL) yöntemleri kullanılmıştır. CBOW temsili ile XGBoost %96,6 doğruluk (Acc), %96,2 F1 skoru sağlayarak en başarılı kombinasyon olmuş ve litaratürle yarışan sonuçlar sonuçlar sergilemiştir. İkinci aşamada, metinler bu kez "cost", "time" ve "quality" boyutlarında etiketlenmiş ve aynı teknik altyapıyla bu yönetimsel temaların sınıflandırılması hedeflenmiştir. Skip-Gram + XGBoost kombinasyonu %88,6 F1 skoruna ulaşsa da, maliyet ve süre ifadelerinin iç içeliği model performansında kısmi belirsizliklere yol açmıştır. Bu durum, daha güçlü bağlamsal analiz araçlarının gerekliliğini ortaya koymuştur. Bu noktadan hareketle, üçüncü aşamada, dört farklı dönüştürücü (Transformer) tabanlı dil modeli (BERT, alBERT, RoBERTa, DistilBERT) ile "GRU, LSTM ve RNN" gibi sıralı derin öğrenme (DL) mimarileri entegre edilmiştir. İki ayrı çok sınıflı sınıflandırma görevinde RoBERTa + LSTM %98,06 Acc ile üstünlük sağlamış, bağlamsal anlam ayrımı gerektiren görevlerde DL modellerin açık başarısı görülmüştür. Elde edilen bulgular, standart inşaat sözleşmelerinin otomatik, bağlamsal ve güvenilir biçimde analizine yönelik yapay zekâ temelli yaklaşımların yüksek başarı potansiyeline sahip olduğunu göstermektedir. Tez kapsamında geliştirilen model, çoklu belge yapılarında anlamlı sınıflandırmalar üretebilmekte; bu yönüyle sözleşme yönetimi, teklif analizi ve risk değerlendirme gibi karar destek süreçlerinde kullanılabilir nitelik taşımaktadır. Literatürde sınırlı yer bulan bağlamsal modellere dair kapsamlı karşılaştırmalar gerçekleştirilmiş, inşaat sektörüne özgü dijital dönüşüm hedeflerine katkı sunulmuştur. Bu sayede, proje yöneticileri ve hukukçular, sözleşme metinlerini analiz edebilir ve hızlı geri bildirimler alarak karar alma süreçlerini daha etkin biçimde yürütebilir.
In this dissertation, a holistic classification framework based on natural language processing (NLP), text mining, and machine learning (ML) techniques has been developed to make sense of the multidimensional content structure of standard Design-Build contracts. Focusing on the Design-Build project delivery method, the research was conducted on the standard English contract texts of the 'FIDIC Conditions of Contract for Design, Build and Operate' and the 'JCT Design and Build Contract'. The contract provisions were classified along both thematic (obligations, tasks, discretionary expressions) and functional (cost, time, quality) dimensions. The study is built upon a three-stage experimental design. In the first stage, contract provisions were categorized into thematic classes such as "obligations," "to do," "optionals," and "other general statements". Classical text representation methods (TF-IDF, BoW, Word2Vec: CBoW & Skip-Gram) were employed in combination with fundamental Machine Learning (ML) algorithms such as "Support Vector Machines (SVM), Logistic Regression (LR), Naive Bayes (NB), K-Nearest Neighbors (KNN), and Decision Trees (DT)", as well as Ensemble Learning (EL) methods including "XGBoost, AdaBoost, Hard Voting, Soft Voting, and Stacking". The combination of CBoW representation and XGBoost yielded the highest performance with 96.6% accuracy (Acc) and 96.2% F1 score, demonstrating results competitive with the literature. In the second stage, the texts were labeled along the dimensions of "cost," "time," and "quality," and the classification of these managerial themes was targeted using the same technical framework. Although the Skip-Gram + XGBoost combination achieved an F1 score of 88.6%, the intertwinement of cost and time expressions introduced partial uncertainties in model performance. This highlighted the need for more robust contextual analysis tools. Building on this, in the third stage, four different transformer-based language models (BERT, alBERT, RoBERTa, DistilBERT) were integrated with sequential Deep Learning (DL) architectures such as "GRU, LSTM, and RNN". In two separate multi-class classification tasks, the RoBERTa + LSTM combination achieved superior performance with 98.06% Acc, clearly demonstrating the success of DL models in tasks requiring contextual semantic distinction. The findings indicate that artificial intelligence-based approaches have a high potential for the automatic, contextual, and reliable analysis of standard construction contracts. The model developed within the scope of this dissertation is capable of producing meaningful classifications across multi-document structures; in this respect, it holds applicability for decision-support processes such as contract management, bid analysis, and risk assessment. Comprehensive comparisons of contextual models sparsely addressed in the literature have been conducted, thereby contributing to the digital transformation goals specific to the construction industry. This allows project managers and legal professionals to interact with contract texts, receive timely feedback, and enhance the efficiency of their decision-making processes.
In this dissertation, a holistic classification framework based on natural language processing (NLP), text mining, and machine learning (ML) techniques has been developed to make sense of the multidimensional content structure of standard Design-Build contracts. Focusing on the Design-Build project delivery method, the research was conducted on the standard English contract texts of the 'FIDIC Conditions of Contract for Design, Build and Operate' and the 'JCT Design and Build Contract'. The contract provisions were classified along both thematic (obligations, tasks, discretionary expressions) and functional (cost, time, quality) dimensions. The study is built upon a three-stage experimental design. In the first stage, contract provisions were categorized into thematic classes such as "obligations," "to do," "optionals," and "other general statements". Classical text representation methods (TF-IDF, BoW, Word2Vec: CBoW & Skip-Gram) were employed in combination with fundamental Machine Learning (ML) algorithms such as "Support Vector Machines (SVM), Logistic Regression (LR), Naive Bayes (NB), K-Nearest Neighbors (KNN), and Decision Trees (DT)", as well as Ensemble Learning (EL) methods including "XGBoost, AdaBoost, Hard Voting, Soft Voting, and Stacking". The combination of CBoW representation and XGBoost yielded the highest performance with 96.6% accuracy (Acc) and 96.2% F1 score, demonstrating results competitive with the literature. In the second stage, the texts were labeled along the dimensions of "cost," "time," and "quality," and the classification of these managerial themes was targeted using the same technical framework. Although the Skip-Gram + XGBoost combination achieved an F1 score of 88.6%, the intertwinement of cost and time expressions introduced partial uncertainties in model performance. This highlighted the need for more robust contextual analysis tools. Building on this, in the third stage, four different transformer-based language models (BERT, alBERT, RoBERTa, DistilBERT) were integrated with sequential Deep Learning (DL) architectures such as "GRU, LSTM, and RNN". In two separate multi-class classification tasks, the RoBERTa + LSTM combination achieved superior performance with 98.06% Acc, clearly demonstrating the success of DL models in tasks requiring contextual semantic distinction. The findings indicate that artificial intelligence-based approaches have a high potential for the automatic, contextual, and reliable analysis of standard construction contracts. The model developed within the scope of this dissertation is capable of producing meaningful classifications across multi-document structures; in this respect, it holds applicability for decision-support processes such as contract management, bid analysis, and risk assessment. Comprehensive comparisons of contextual models sparsely addressed in the literature have been conducted, thereby contributing to the digital transformation goals specific to the construction industry. This allows project managers and legal professionals to interact with contract texts, receive timely feedback, and enhance the efficiency of their decision-making processes.
Açıklama
Anahtar Kelimeler
İnşaat Mühendisliği
Kaynak
WoS Q Değeri
Scopus Q Değeri
Cilt
Sayı
Künye
Demircan, A. (2025). Metin Madenciliği Ve Makine Öğrenimi Tekniklerinin Uygulanmasıyla Standart İnşaat Sözleşmesi Belgelerinin Metin Analizi (Tez No. 968429) [Doktora Tezi, Düzce Üniversitesi].












