Single-Document Summarization Using Sentence Embeddings and K-Means Clustering

Agarwal, Seema; Singh, Nikhil Kumar; Meel, Priyanka

doi:10.1109/icacccn.2018.8748762

Cited by 8 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…What types of methods have been used for AESDS recently? A significant group of 13 studies (54%) based their proposals on machine learning (ML) [7]- [13] or hybridizing neural networks with techniques such as graphs [14], metaheuristics [15], differential evolution [16], NLP [17], [18], or genetic algorithms [19], and most of the methods present in the ranking of state of the art are among these proposals.…”

Section: Resultsmentioning

confidence: 99%

“…In the studies reviewed, one feature or the combination of several features (naive multi-objective approach) can be evaluated in a single objective. From these, the most commonly used means [13], [20], [21]; K-medoids [15], [22], [23]; neural network-based clustering, e.g., self-organizing maps (SOM) [7]- [9], [15], [16]; topic detection such as LDA [14]; clustering by keyword identification [11], [18], [24] or using non-negative factorization matrices [12], [25].…”

Section: Q3 What Characteristics Are Considered For Aesds?mentioning

confidence: 99%

See 1 more Smart Citation

Generación automática de resúmenes extractivos para un solo documento: un mapeo sistemático

Yip-Herrera,

Mendoza-Becerra,

Rodríguez

2023

Rev. Fac. Ing.

View full text Add to dashboard Cite

La Generación Automática de Resúmenes Extractivos para un Solo Documento (GAReUD) es un área de investigación que tiene como objetivo crear una versión corta de un documento con la información más relevante y adquiere mayor importancia a diario debido a la necesidad de los usuarios de obtener rápidamente información de documentos publicados en internet. En el área de generación automática de resúmenes cada elemento debe ser evaluado y luego rankeado para conformar un resumen, de acuerdo con esto, existen tres diferentes enfoques teniendo en cuenta la cantidad de objetivos que se evalúan, así: mono objetivo, multi objetivo y de muchos objetivos. El propósito de este mapeo sistemático es brindar conocimiento sobre los métodos y técnicas utilizadas en métodos extractivos de GAReUD, analizando la cantidad de objetivos y características evaluadas, que pueden ser útiles para futuras investigaciones. Este mapeo se realizó utilizando un proceso genérico para la realización de revisiones sistemáticas donde se construye una cadena de búsqueda considerando unas preguntas de investigación, luego se utiliza un filtro con unos criterios de inclusión y exclusión para la selección de los estudios primarios con los que se realizará el análisis, adicionalmente, estos estudios se ordenan de acuerdo con la relevancia de su contenido; este proceso se resume en tres pasos principales: Planificación, Ejecución y Análisis de resultados. Al final del mapeo se identificaron las siguientes observaciones: (i) existe una preferencia por la utilización de métodos basados en aprendizaje automático de máquina y también por el uso de técnicas de agrupamiento, (ii) la importancia de usar como objetivos ambos tipos de características (estadísticas y semánticas) y (iii) la necesidad de explorar el enfoque de muchos objetivos.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Q3 What Characteristics Are Considered For Aesds?mentioning

confidence: 99%

Generación automática de resúmenes extractivos para un solo documento: un mapeo sistemático

Yip-Herrera,

Mendoza-Becerra,

Rodríguez

2023

Rev. Fac. Ing.

View full text Add to dashboard Cite

show abstract

“…The sentence embeddings were clustered as per the summary size requirements using K-means method. Thereafter, 'Ridge Regression' sentence scoring model was used to pick the most relevant sentences from each cluster to formulate a summary [15].…”

Section: Related Workmentioning

confidence: 99%

Unsupervised Automatic Text Summarization of Konkani Texts using K-means with Elbow Method

D’Silva¹,

Sharma²

2020

International Journal of Engineering Research and Technology

View full text Add to dashboard Cite

Text Summarization is an emerging field of research in Natural Language Processing (NLP). A bulk of the work is related to texts in English and other popular languages. This paper presents some of the early works attempted at performing single document extractive Automatic Text Summarization on Konkani language documents, which is an under-research language in the domain of Automatic Text Summarization (ATS). The input documents need to be cleaned of punctuation and then sentence scores are calculated for each sentence in the document. The scores for each sentence are computed using Term-Frequency/Inverse Document Frequency (TF-IDF) of constituent words and overlap with the title of the story and its positional value. K-means algorithm is applied to determine clusters of sentences for the formation of the final summary. The value of 'K' is determined using the Elbow method. The dataset employed was specially designed by the authors of the paper to perform the experiments. It consists of folk tales derived from books on Konkani literature. The performance assessment of the output summaries indicated that the summaries obtained by using three clusters were better than the ones obtained using two clusters. The proposed system exhibited promising outcome, considering, no language-dependent domain knowledge or any training corpora was utilized.

show abstract

“…The emergence of deep learning [2][3][4] [5] models has driven the advancement of text summarization technology. For instance, Padmakumar [6] and others employed clustering methods [7][8] [9] to generate text summaries, dividing sentences into multiple categories and selecting sentences closest to the center as the summary. Zhang [10] and colleagues combined Seq2seq models with sequence labeling methods to compute probability distributions and obtain document summaries through sampling.…”

Section: Introductionmentioning

confidence: 99%

Automatic audit system based on intelligent mining of audit work order text

Zhao,

Liu,

et al. 2024

Third International Conference on Electronic Information Engineering, Big Data, and Computer Technology (EIBDCT 2024)

View full text Add to dashboard Cite

With the rapid development of electricity marketing business, the volume of business and data for power grid companies has significantly increased. Traditional manual auditing methods are unable to meet modern management demands. According to statistics, the manual misjudgment rate during the quality inspection process of audit work orders has reached 17.8%. This article establishes an intelligent audit model for audit work orders based on text mining technology. Firstly, pre-trained models are used to generate sentence vectors. Then, TextRank algorithm and MMR algorithm are integrated to generate summaries. Subsequently, using a Transformer model with incorporated sparse attention mechanism, the generated summaries are used as inputs for binary classification to obtain audit results. Experimental results demonstrate that the proposed audit model for work orders can achieve automated processing and fine-grained analysis of marketing audit monitoring tasks, effectively improving the efficiency and accuracy of the auditing process.

show abstract

Single-Document Summarization Using Sentence Embeddings and K-Means Clustering

Cited by 8 publications

References 7 publications

Generación automática de resúmenes extractivos para un solo documento: un mapeo sistemático

Generación automática de resúmenes extractivos para un solo documento: un mapeo sistemático

Unsupervised Automatic Text Summarization of Konkani Texts using K-means with Elbow Method

Automatic audit system based on intelligent mining of audit work order text

Contact Info

Product

Resources

About