Extraction and Identification of Bar Graph Components by Automatic Epsilon Estimation

Kanjanawattana, Sarunya; Kimura, Masaomi

doi:10.7763/ijcte.2017.v9.1147

Cited by 2 publications

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main idea of this research was to propose methods extracting information from graphical and linguistic representations as well as utilize them to express explicit and implicit knowledge. We used the method of graph component extraction [5] and graph information extraction [6] to extract information from the graphical data, including graphs' linguistic data, e.g. caption.…”

Section: Discussionmentioning

confidence: 99%

“…We utilize graph component extraction [5], graph information extraction [6], and optical-characterrecognition-error (OCR-error) correction [25] to obtain the graph information. This graph information is then added to an ontology with the structure designed in the previous study [6].…”

Section: Methodsmentioning

confidence: 99%

“…In the past, we have attempted to extract graphical information from a graph, including graph components such as axis titles and legends [5] and graph data information such as data located in the axes of the graph [6], [7]. We constructed an ontology and stored the information extracted from the graphs and their descriptive contents (i.e., captions and cited paragraphs) in the ontology.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Semantic-Based Search Engine System for Graph Images in Academic Literatures by Use of Semantic Relationships

Kanjanawattana

Kimura²

2019

IJMLC

Self Cite

View full text Add to dashboard Cite

Information retrieval is a baseline of search engine systems. There is a very large amount of data published on the Internet that cannot be manually searched. However, search engine systems should not only present relevant results but also obtain new knowledge from the user's searches. For example, new knowledge in academic research areas may be presented in graph images. In this study, we utilize methods to extract graphical and textual information from graph images and store this new knowledge in an ontology. We propose a search engine system that is applicable to an ontology that contains this extractable information, which is extracted from images with graphs. The developed ontology is useful because users can acquire a considerable amount of knowledge that is discovered from the semantic relations in the ontology. To evaluate the search engine system, ten participants tested the system and responded to their feedback. The results indicate that the proposed system provides accurate and relevant results; moreover, as indicated by the higher F-measure values comparing to an Elasticsearch-based search engine system, the performance of our system is highly acceptable. It clarified that the ontology-based search engine system provides precise and concise information outperforming than the Elasticsearchbased search engine system.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%