Dataset Information

Reaching for upper bound ROUGE score of extractive summarization methods.

ABSTRACT: The extractive text summarization (ETS) method for finding the salient information from a text automatically uses the exact sentences from the source text. In this article, we answer the question of what quality of a summary we can achieve with ETS methods? To maximize the ROUGE-1 score, we used five approaches: (1) adapted reduced variable neighborhood search (RVNS), (2) Greedy algorithm, (3) VNS initialized by Greedy algorithm results, (4) genetic algorithm, and (5) genetic algorithm initialized by the Greedy algorithm results. Furthermore, we ran experiments on articles from the arXive dataset. As a result, we found 0.59 and 0.25 scores for ROUGE-1 and ROUGE-2, respectively achievable by the approach, where the genetic algorithm initialized by the Greedy algorithm results, which happens to yield the best results out of the tested approaches. Moreover, those scores appear to be higher than scores obtained by the current state-of-the-art text summarization models: the best score in the literature for ROUGE-1 on the same data set is 0.46. Therefore, we have room for the development of ETS methods, which are now undeservedly forgotten.

SUBMITTER: Akhmetov I

PROVIDER: S-EPMC9575858 | biostudies-literature | 2022

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Reaching for upper bound ROUGE score of extractive summarization methods.

Akhmetov Iskander I Mussabayev Rustam R Gelbukh Alexander A

PeerJ. Computer science 20220926

The extractive text summarization (ETS) method for finding the salient information from a text automatically uses the exact sentences from the source text. In this article, we answer the question of what quality of a summary we can achieve with ETS methods? To maximize the ROUGE-1 score, we used five approaches: (1) adapted reduced variable neighborhood search (RVNS), (2) Greedy algorithm, (3) VNS initialized by Greedy algorithm results, (4) genetic algorithm, and (5) genetic algorithm initializ ...[more]

PMID: 36262160

Similar Datasets

Project description:Automated summarization of clinical texts can reduce the burden of medical professionals. "Discharge summaries" are one promising application of the summarization, because they can be generated from daily inpatient records. Our preliminary experiment suggests that 20-31% of the descriptions in discharge summaries overlap with the content of the inpatient records. However, it remains unclear how the summaries should be generated from the unstructured source. To decompose the physician's summarization process, this study aimed to identify the optimal granularity in summarization. We first defined three types of summarization units with different granularities to compare the performance of the discharge summary generation: whole sentences, clinical segments, and clauses. We defined clinical segments in this study, aiming to express the smallest medically meaningful concepts. To obtain the clinical segments, it was necessary to automatically split the texts in the first stage of the pipeline. Accordingly, we compared rule-based methods and a machine learning method, and the latter outperformed the formers with an F1 score of 0.846 in the splitting task. Next, we experimentally measured the accuracy of extractive summarization using the three types of units, based on the ROUGE-1 metric, on a multi-institutional national archive of health records in Japan. The measured accuracies of extractive summarization using whole sentences, clinical segments, and clauses were 31.91, 36.15, and 25.18, respectively. We found that the clinical segments yielded higher accuracy than sentences and clauses. This result indicates that summarization of inpatient records demands finer granularity than sentence-oriented processing. Although we used only Japanese health records, it can be interpreted as follows: physicians extract "concepts of medical significance" from patient records and recombine them in new contexts when summarizing chronological clinical records, rather than simply copying and pasting topic sentences. This observation suggests that a discharge summary is created by higher-order information processing over concepts on sub-sentence level, which may guide future research in this field.

Project description:BACKGROUND:Doctors must care for many patients simultaneously, and it is time-consuming to find and examine all patients' medical histories. Discharge diagnoses provide hospital staff with sufficient information to enable handling multiple patients; however, the excessive amount of words in the diagnostic sentences poses problems. Deep learning may be an effective solution to overcome this problem, but the use of such a heavy model may also add another obstacle to systems with limited computing resources. OBJECTIVE:We aimed to build a diagnoses-extractive summarization model for hospital information systems and provide a service that can be operated even with limited computing resources. METHODS:We used a Bidirectional Encoder Representations from Transformers (BERT)-based structure with a two-stage training method based on 258,050 discharge diagnoses obtained from the National Taiwan University Hospital Integrated Medical Database, and the highlighted extractive summaries written by experienced doctors were labeled. The model size was reduced using a character-level token, the number of parameters was decreased from 108,523,714 to 963,496, and the model was pretrained using random mask characters in the discharge diagnoses and International Statistical Classification of Diseases and Related Health Problems sets. We then fine-tuned the model using summary labels and cleaned up the prediction results by averaging all probabilities for entire words to prevent character level-induced fragment words. Model performance was evaluated against existing models BERT, BioBERT, and Long Short-Term Memory (LSTM) using the Recall-Oriented Understudy for Gisting Evaluation (ROUGE) L score, and a questionnaire website was built to collect feedback from more doctors for each summary proposal. RESULTS:The area under the receiver operating characteristic curve values of the summary proposals were 0.928, 0.941, 0.899, and 0.947 for BERT, BioBERT, LSTM, and the proposed model (AlphaBERT), respectively. The ROUGE-L scores were 0.697, 0.711, 0.648, and 0.693 for BERT, BioBERT, LSTM, and AlphaBERT, respectively. The mean (SD) critique scores from doctors were 2.232 (0.832), 2.134 (0.877), 2.207 (0.844), 1.927 (0.910), and 2.126 (0.874) for reference-by-doctor labels, BERT, BioBERT, LSTM, and AlphaBERT, respectively. Based on the paired t test, there was a statistically significant difference in LSTM compared to the reference (P<.001), BERT (P=.001), BioBERT (P<.001), and AlphaBERT (P=.002), but not in the other models. CONCLUSIONS:Use of character-level tokens in a BERT model can greatly decrease the model size without significantly reducing performance for diagnoses summarization. A well-developed deep-learning model will enhance doctors' abilities to manage patients and promote medical studies by providing the capability to use extensive unstructured free-text notes.

Dataset Information

Reaching for upper bound ROUGE score of extractive summarization methods.

Publications

Reaching for upper bound ROUGE score of extractive summarization methods.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets