Search

Copyright of Journal of Engineering & Architectural Faculty of Eskisehir Osmangazi University / Eskişehir Osmangazi Üniversitesi Mühendislik ve Mimarlık Fakültesi Dergisi is the property of Eskisehir Osmangazi University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Text Summarization models facilitate biomedical clinicians and researchers in acquiring informative data from enormous domain-specific literature within less time and effort. Evaluating and selecting the most informative sentences from biomedical articles is always challenging. This study aims to develop a dual-mode biomedical text summarization model to achieve enhanced coverage and information. The research also includes checking the fitment of appropriate graph ranking techniques for improved performance of the summarization model. The input biomedical text is mapped as a graph where meaningful sentences are evaluated as the central node and the critical associations between them. The proposed framework utilizes the top k similarity technique in a combination of UMLS and a sampled probability-based clustering method which aids in unearthing relevant meanings of the biomedical domain-specific word vectors and finding the best possible associations between crucial sentences. The quality of the framework is assessed via different parameters like information retention, coverage, readability, cohesion, and ROUGE scores in clustering and non-clustering modes. The significant benefits of the suggested technique are capturing crucial biomedical information with increased coverage and reasonable memory consumption. The configurable settings of combined parameters reduce execution time, enhance memory utilization, and extract relevant information outperforming other biomedical baseline models. An improvement of 17% is achieved when the proposed model is checked against similar biomedical text summarizers. [ABSTRACT FROM AUTHOR]

A novel approach to the semantic comparison of texts based on the metric method to calculate semantic distances between lexical units is put forward. The supplementary semantic information is provided through semes of the words composing the texts, or through their semantic fields. The proposed method takes into account semantic polarity and yields, for two paraphrase sentences, more feasible results than the conventional approaches based on word occurrences. The described approach may be useful for linguistic theory as well as for a variety of Natural Language Processing tasks based on supervised learning that require semantic information: computer lexicography, semantic analysis, information search and retrieval, document classification, text summarization, and understanding machine translation and others. [ABSTRACT FROM AUTHOR]

Text summarization is the process of reducing text size while preserving its key points. Thanks to this process, the reading time of the text is also reduced which contributes to reaching the desired information quickly, especially in today's world where time is much more important. In addition, summarization can be used to create a solution to extract outstanding information from the text. In this study, we focus on abstract summarization, which can draw more human like conclusions from the text. A summarization study was carried out on the data set that was collected from online Turkish news sources. Rouge and Bert-score performance metrics were used to compare the performance of this study using the text to text transfer transformer (T5) method. The precision values ​​of the Rouge-1, Rouge-2, Rouge-L and Bert-score performance metrics obtained in this study were found to be 0.6913, 0.6623, 0.7528 and 0.8718, respectively. Recall values ​​were 0.9210, 0.8917, 0.9183 and 0.9138, respectively. F measure values ​​were 0.7649, 0.7338, 0.8084 and 0.8913 respectively. Considering the success of the results obtained in the study, a method that can obtain successful results for Turkish text summarization is presented and the original dataset is made available to other researchers. [ABSTRACT FROM AUTHOR]

Text classification is very useful to search large amount of textual data available online by dividing it into smaller relevant units. Now a day's large amount of digital documents are available in Indian languages. Designing text classifiers in Indian languages is one of the research areas so that people can search and read required documents in their local languages. In proposed work tried to design Text classifier for Hindi text documents and tried to show how stemmer affects the performance of Hindi text classifiers. Stemming is a process to convert words in any language to its base or root words. Stemmers are used for written documents not for spoken languages. Performance of many applications such as text summarization, Information Retrieval (IR) system, text classification systems, syntactic parsing can be improved by applying stemmers. Stemmer eliminates suffix or prefix of the word and form original root word. These root words helps in the preprocessing step required in many algorithms. We applied various stemmers on Hindi text classification models. Experiments and results show that performance of the classifiers is improved by applying stemmers. [ABSTRACT FROM AUTHOR]

Conditional text generation is supposed to generate a fluent and coherent target text that is faithful to the source text. Although pre-trained models have achieved promising results, they still suffer from the crucial factuality problem. To deal with this issue, we propose a factuality-aware pretraining-finetuning framework named FactGen, which fully considers factuality during two training stages. Specifically, at the pre-training stage, we utilize a natural language inference model to construct target texts that are entailed by the source texts, resulting in a more factually consistent pre-training objective. Then, during the fine-tuning stage, we further introduce a contrastive ranking loss to encourage the model to generate factually consistent text with higher probability. Extensive experiments on three conditional text generation tasks demonstrate the effectiveness and generality of our training framework. [ABSTRACT FROM AUTHOR]

With the rapid advancement in healthcare, there has been exponential growth in the healthcare records stored in large databases to help researchers, clinicians, and medical practitioner's for optimal patient care, research, and trials. Since these studies and records are lengthy and time consuming for clinicians and medical practitioners, there is a demand for new, fast, and intelligent medical information retrieval methods. The present study is a part of the project which aims to design an intelligent medical information retrieval and summarization system. The whole system comprises three main modules, namely adverse drug event classification (ADEC), medical named entity recognition (MNER), and multi-model text summarization (MMTS). In the current study, we are presenting the design of the ADEC module for classification tasks, where basic machine learning (ML) and deep learning (DL) techniques, such as logistic regression (LR), decision tree (DT), and text-based convolutional neural network (TextCNN) are employed. In order to perform the extraction of features from the text data, TF-IDF and Word2Vec models are employed. To achieve the best performance of the overall system for efficient information retrieval and summarization, an ensemble strategy is employed, where predictions of the selected base models are integrated to boost the robustness of one model. The performance results of all the models are recorded as promising. TextCNN, with an accuracy of 89%, performs better than the conventional machine learning approaches, i.e., LR and DT with accuracies of 85% and 77%, respectively. Furthermore, the proposed TextCNN outperforms the existing adverse drug event classification approaches, achieving precision, recall, and an F1 score of 87%, 91%, and 89%, respectively. [ABSTRACT FROM AUTHOR]

Automated information retrieval and text summarization concept is a difficult process in natural language processing because of the infrequent structure and high complexity of the documents. The text summarization process creates a summary by paraphrasing a long text. Earlier models on information retrieval and summarization are based on a massive labeled dataset by the use of handcrafted features, leveraging on knowledge for a particular domain, and concentrated on the narrow sub-domain to improve efficiency. This paper presents a new deep learning (DL) based information retrieval with a text summarization model. The proposed model involves three major processes namely information retrieval, template generation, and text summarization. Initially, the bidirectional long short term memory (BiLSTM) approach is employed for retrieving the textual data, which assumes each word in a sentence, extracts the information, and embeds it into the semantic vector. Next, the template generation process takes place using the DL model. The deep belief network (DBN) model is employed as a text summarization tool to summarize the textual content. In addition, the image description is generated for the visualized entities that exist in the images. The design of BiLSTM with the DBN model for the text summarization and image captioning process shows the novelty of the work. The performance of the presented method is validated using Giga word corpus and DUC corpus. The experimental results referred that the proposed DBN model outperformed the compared methods with the maximum precision, recall and F-score. The image captions are compared with a predefined set of captions that exists for the image and the performance is evaluated using the BLEU metric.

With the rapid development of the Internet, the massive amount of web textual data has grown exponentially, which has brought considerable challenges to downstream tasks, such as document management, text classification, and information retrieval. Automatic text summarization (ATS) is becoming an extremely important means to solve this problem. The core of ATS is to mine the gist of the original text and automatically generate a concise and readable summary. Recently, to better balance and develop these two aspects, deep learning (DL)-based abstractive summarization models have been developed. At present, for ATS tasks, almost all state-of-the-art (SOTA) models are based on DL architecture. However, a comprehensive literature survey is still lacking in the field of DL-based abstractive text summarization. To fill this gap, this paper provides researchers with a comprehensive survey of DL-based abstractive summarization. We first give an overview of abstractive summarization and DL. Then, we summarize several typical frameworks of abstractive summarization. After that, we also give a comparison of several popular datasets that are commonly used for training, validation, and testing. We further analyze the performance of several typical abstractive summarization systems on common datasets. Finally, we highlight some open challenges in the abstractive summarization task and outline some future research trends. We hope that these explorations will provide researchers with new insights into DL-based abstractive summarization. [ABSTRACT FROM AUTHOR]

Exponential increase in the amount of textual data made available on the Internet results in new challenges in terms of accessing information accurately and quickly. Text summarization can be defined as reducing the dimensions of the expressions to be summarized without spoiling the meaning. Summarization can be performed as extractive and abstractive or using both together. In this study, we focus on abstractive summarization which can produce more human‐like summarization results. For the study we created a Turkish news summarization benchmark dataset from various news agency web portals by crawling the news title, short news, news content, and keywords for the last 5 years. The dataset is made publicly available for researchers. The deep learning network training was carried out by using the news headlines and short news contents from the prepared dataset and then the network was expected to create the news headline as the short news summary. To evaluate the performance of this study, Rouge‐1, Rouge‐2, and Rouge‐L were compared using precision, sensitivity and F1 measure scores. Performance values for the study were presented for each sentence as well as by averaging the results for 50 randomly selected sentences. The F1 Measure values are 0.4317, 0.2194, and 0.4334 for Rouge‐1, Rouge‐2, and Rouge‐L respectively. Performance results show that the approach is promising for Turkish text summarization studies and the prepared dataset will add value to the literature. [ABSTRACT FROM AUTHOR]

The notion of Textual Entailment (TE) is an established indicator of text connectedness. It captures semantic relationships between texts. Recently, it has been used successfully for determining sentence salience in many text summarization methods. However, it has been reported in previous works that the standard textual entailment is not ideal for measuring sentence salience. This is because textual entailment relationships between sentences are quite rare in real-world texts. Therefore, we suggest using partial TE to accomplish the task of recognizing standard TE. We present the single document summarization problem as an optimization problem which is solved using a weighted Minimum Set Cover (wMSC) algorithm. In this method, sentences are broken into fragments and Partial TE is used to form sets of fragments. Finally, wMSC is applied to the sets to obtain the minimum set cover, which corresponds to the summary of the document. The results achieved on the DUC 2002 dataset using ROUGE and other quality metrics show that the proposed method outperforms the state of the art. [ABSTRACT FROM AUTHOR]

With the advent of transformer-based architectures, the contextual representation of text data has leveraged the query and the document to be represented in low-dimensional dense vector space. These vectors are learned embeddings of fixed sizes, resulting in deeper text understanding. In this study, we designed a pipeline for effectively retrieving documents from a large search space by combining the deeper text understanding capabilities of the transformer-based BERT model and a phrase embedding-based query expansion model. To learn the contextual representations, we fine-tuned a deep semantic matching model by separately encoding the document and the query. The encoder model is based on the Sentence BERT (SBERT) architecture, which separately generates dense vector representations of documents and queries. The study has also addressed the maximum token length limitation of transformer-based models through the summarization of lengthy documents. In addition, to improve the clarity and completeness of short queries and reduce the semantic gap, a phrase embedding-based query expansion model is employed. The documents and their dense vectors are indexed using the Elasticsearch engine, and matched them with query vectors for retrieving query-specific documents. Finally, the BERT-based cross-encoder model is used to re-rank the relevant records for each query. It performs full self-attention over the inputs, and yields richer text interactions to produce the final results. To assess performance, experiments are conducted on two well-known datasets, TREC-CDS-2014 and OHSUMED. A comparative analysis is carried out, which clearly demonstrates that the proposed framework produced competitive retrieval results. • Transformer-based contextual text representation approach is proposed for IR. • Bi-Encoder architecture is applied to learn the representations of documents & query. • BERT-based summarization is applied to address the limitation on sequence length. • Phrase Embedding-based QE technique is employed to expand short and unclear queries. • Re-ranking of the retrieval results is performed using the cross encoder model. [ABSTRACT FROM AUTHOR]

Development of effective automatic summarization approaches for legal documents suffer from several challenges like extremely long document-summary pairs, lack of large scale training datasets with tractable document-summary token lengths, etc. In this work, we deal with the problem of legal document summarization by building a modified abstractive dataset from the original dataset. This ensures that the length of each document-summary pair is manageable and can be processed by the state of the art summarization approaches (such as BART). Secondly, we deal with the data scarcity problem by creating more number of training samples, from each of the original document-summary pair. This is done by creating multiple extractive summaries from each sample in the original dataset, following which ground-truth summary sentences are assigned to each of the extractive summary to generate new training samples. This results in a larger training dataset that can be utilized for fine-tuning summarization models. Our proposed approach has been evaluated on two different legal datasets- BillSum and Forum of Information Retrieval Evaluation (FIRE). With respect to the ROUGE metrics, the proposed approach is able to outperform pre-trained BART model fine-tuned on original dataset by (3 − 8) % for FIRE test sets, and by (1 − 3) % for the BillSum test sets. Considering the BERTScore metrics, the proposed approach obtains (1 − 2) % improvements on the FIRE test sets, while for the BillSum test sets (3 − 8) % improvements are observed. Such improvements suggest that the proposed dataset building approach can help achieve improved abstractive summarization of lengthy legal documents. • Lengthy nature and data scarcity are the two main challenges with legal documents. • Lengthy legal document summarization is handled by the proposed approach. • The data scarcity problem is handled by creating a feasible abstractive dataset. • A novel Extract-Then-Assign (ETA) approach is proposed. • ETA approach can greatly improve abstractive summarization of legal documents. [ABSTRACT FROM AUTHOR]

The World Wide Web is an ever expanding source of data in today's world. Millions of tera-bytes of data and information is getting added every second. In this information age as the data is getting generated at an exponential rate, the fact to be noted is that most of the information is already available is in the form of natural language text. The task of information extraction from mammoth data leads us to think on the quality and the form of available data. Secondly, the ever increasing data poses a challenging task of extracting useful information from the available data. The third task is to extract information as efficiently as possible. For retrieving the information there is a need to develop ingenious way to answer any kind of query put up by a user from the available unstructured data. This paper proposes a novel trainable and integrated Natural Language Information Interpretation and Representation System (NLIIRS) that accepts any available un-annotated corpus of data in the form of natural language, and performs the following tasks: finds out the useful data, extracts relevant information in usable form (structured form/tables), summarizes the data and structures the data in relational form. At the end the Question and Answering (Q&A) module shows the cognitive abilities of NLIIR system by answering the questions in natural language relevant to the text. This multispecialty system beyond just Q&A. This is a trainable system capable of handling any unstructured data to be transformed into structured and well organized information. It allows the user to ask questions in natural language. It adopts the advantages of a modified named entity recognition so as to bypass the time consuming process of parts of speech tagging while pre-processing the available corpus (data) for information extraction. [ABSTRACT FROM AUTHOR]

Automatic text summarization is a field situated at the intersection of natural language processing and information retrieval. Its main objective is to automatically produce a condensed representative form of documents. This paper presents ArA*summarizer, an automatic system for Arabic single document summarization. The system is based on an unsupervised hybrid approach that combines statistical, cluster‐based, and graph‐based techniques. The main idea is to divide text into subtopics then select the most relevant sentences in the most relevant subtopics. The selection process is done by an A* algorithm executed on a graph representing the different lexical–semantic relationships between sentences. Experimentation is conducted on Essex Arabic summaries corpus and using recall‐oriented understudy for gisting evaluation, automatic summarization engineering, merged model graphs, and n‐gram graph powered evaluation via regression evaluation metrics. The evaluation results showed the good performance of our system compared with existing works. [ABSTRACT FROM AUTHOR]

Automatic text summarization is a process of condensing the content of a text document to pursue the most important information. It plays a significant role in various tasks like text categorization, question answering and information retrieval (IR). As legal information retrieval (LIR) is a subfield of IR, the produced summaries are combined into IR system, with the objective of decreasing the length of the document. In this way, we can improve the access time for searching the information, and relevant documents are retrieved. In this article, we present the creation of passage-level summaries (generic and legal) with different compression ratios and evaluate their performance. The generic summaries present the overall description of the essential information of a document and legal summaries, produced by taking into account the domain-specific features that are present in the document. Next, we propose Boosting Okapi BM25 which is the modified model of Okapi BM25 to increase the efficiency of the LIR. We have evaluated proposed LIR approach in terms of MAP and R-precision and summarization approach using ROUGE tool on FIRE2013 and FIRE2014 datasets. To show the efficacy of the proposed system, we compare the experimental results with different IR models like PL2, I n _ e x p B 2 , I n _ e x p C 2 , InL2, D F R _ B M 25 , Okapi BM25 in terms of MAP. The experimental results of the proposed system show better performance than the existing various IR models in terms of various performance metrics. The empirical results also exhibit that the integration of text summarization and IR techniques helps in retrieving relevant information with less access time. [ABSTRACT FROM AUTHOR]

Abstract Part-of-Speech (POS) Tagging is the process of automatically determining the proper grammatical tag or syntactic category of a word depending on a its context. POS Tagging is an essential step in most Natural Language Processing (NLP) applications such as text summarization, question answering, information extraction and information retrieval. In this study, we propose an efficient tagging approach for the Arabic language using Bee Colony Optimization algorithm. The problem is represented as a graph and a novel technique is proposed to assign scores to possible tags of a sentence, then the bees find the best solution path. The proposed approach is evaluated using KALIMAT corpus which consists of 18M words. Experimental results showed that the proposed approach achieved 98.2% of accuracy compared to 98%, 97.4% and 94.6% for Hybrid, Hidden Markov Model and Rule-Based methods respectively. Furthermore, the proposed approach determined all the tags presented in the corpus while the mentioned approaches can identify only three tags. [ABSTRACT FROM AUTHOR]

Text summarization is an application of information retrieval where short and non-redundant version of comparatively large text is presented to the end user. In this paper a hybrid approach is presented to generate abstract summary in which sentences are clustered using sentence level relationships among sentences in association with Markov clustering principle. Then sentence ranking is done in each cluster and if possible the top weighted sentence of each cluster is fused using some linguistic rules with its best-fit sentence(if found) within that cluster to generate a new sentence. Then top ranked sentences from each cluster are compressed using classification technique to generate the abstract summary. The proposed system is evaluated with DUC 2002 data set and the result is performing better than other existing systems. [ABSTRACT FROM AUTHOR]

The availability of various digital sources has created a demand for text mining mechanisms. Effective summary generation mechanisms are needed in order to utilize relevant information from often overwhelming digital data sources. In this view, this paper conducts a survey of various single as well as multi-document text summarization techniques. It also provides analysis of treating a query sentence as a common one, segmented from documents for text summarization. Experimental results show the degree of effectiveness in text summarization over different clustering algorithms. [ABSTRACT FROM AUTHOR]

Stemming is an important pre-processing step in the text analysis domains such as text mining, text summarization and information retrieval (IR). In this study, we build a Sanskrit text collection and explore different indexing, stemming and searching strategies in Sanskrit. We also propose two stemmers: a 'light' and an 'aggressive' and evaluate their effectiveness in the text analysis task. The performance of the stemmers is evaluated in two ways: a direct and an indirect IR-based evaluation. In direct evaluation, we found that the stemmers are effective. In indirect evaluation, we apply different retrieval models such as BM25, TF–IDF, Divergence from Randomness (DFR) based and language models. The proposed stemmers are compared with GRAS stemmer, language-independent indexing (trunc-n) and no stemming approach. Among different stemming methods, aggressive stemmers provide the best performance. Hiemstra language model outperforms other retrieval models we experimented with. In statistical analysis, we found that the proposed stemming approaches produce significantly better results than the no-stemming approach. • We build a Sanskrit text collection for the text analysis domain. • We proposed an inflectional and derivational stemmer in Sanskrit. • The performance of different stemmer is evaluated in the text analysis domain. [ABSTRACT FROM AUTHOR]

Generative Summarization is of great importance in understanding large-scale textual data. In this work, we propose an attention-based Tree-LSTM model for sentence summarization, which utilizes an attention-based syntactic structure as auxiliary information. Thereinto, block-alignment is used to align the input and output syntax blocks, while inter-alignment is used for alignment of words within that of block pairs. To some extent, block-alignment can prevent structural deviations on the long sentences and inter-alignment is capable of increasing the flexibility of the generation in the blocks. This model can be easily trained to end-to-end mode and deal with any length of the input sentences. Compared with several relatively strong baselines, our model has achieved the state-of-art on DUC-2004 shared task. [ABSTRACT FROM AUTHOR]

As information is available in abundance for every topic on internet, condensing the important information in the form of summary would benefit a number of users. Hence, there is growing interest among the research community for developing new approaches to automatically summarize the text. Automatic text summarization system generates a summary, i.e. short length text that includes all the important information of the document. Since the advent of text summarization in 1950s, researchers have been trying to improve techniques for generating summaries so that machine generated summary matches with the human made summary. Summary can be generated through extractive as well as abstractive methods. Abstractive methods are highly complex as they need extensive natural language processing. Therefore, research community is focusing more on extractive summaries, trying to achieve more coherent and meaningful summaries. During a decade, several extractive approaches have been developed for automatic summary generation that implements a number of machine learning and optimization techniques. This paper presents a comprehensive survey of recent text summarization extractive approaches developed in the last decade. Their needs are identified and their advantages and disadvantages are listed in a comparative manner. A few abstractive and multilingual text summarization approaches are also covered. Summary evaluation is another challenging issue in this research field. Therefore, intrinsic as well as extrinsic both the methods of summary evaluation are described in detail along with text summarization evaluation conferences and workshops. Furthermore, evaluation results of extractive summarization approaches are presented on some shared DUC datasets. Finally this paper concludes with the discussion of useful future directions that can help researchers to identify areas where further research is needed. [ABSTRACT FROM AUTHOR]

The information overload problem has posed great challenge to internet users to retrieve relevant information accurately for the past decades. It is a tedious task for machine to intuitively mimic human linguists to summarize documents into meaningful text in abstractive manner. Quite often, the summarized text lacks cohesion and becomes difficult to comprehend. The objective of this paper is to investigate the proposed Swarm LSA-PSO model performs better than alternative methods. In this study, terms matrix was constructed from co-occurrence of terms using Bag-of-Words (BOW). The huge dimensions of terms were reduced using Singular Value Decomposition followed by K-Means PSO clustering for acquiring optimal number of concepts clusters. These key concepts were used to identify the main gist in documents for text summarization. The input text documents were downloaded from Document Understanding Conference (DUC) 2002 dataset. The preliminary results show that the swarm LSA-PSO model shows promising results in context based text summarization using BOW clustering approach. [ABSTRACT FROM AUTHOR]

Computational linguistics encompasses a broad range of ideas and research areas, and only a brief introduction is possible here. We chose to include areas in computational linguistics where statisticians can contribute, hoping to provide inspiration to the reader. We describe three main aspects of this discipline-formal languages, information retrieval, and machine learning. These support the overarching goal, which is the representation and analysis of meaning from unstructured text. We then provide an example where text analysis has been applied to unstructured text fields in survey records and conclude with some applications and computational resources. WIREs Comput Stat 2015, 7:258-274. doi: 10.1002/wics.1353 For further resources related to this article, please visit the . Conflict of interest: The authors have declared no conflicts of interest for this article. [ABSTRACT FROM AUTHOR]