A Comprehensive Evaluation of Cue-Words based Features and In-text Citations based Features for Citation Classification

Citation plays a vital role in the scientific community of evaluating the contributions of scientific authors. Citing sources delivers a measurable way of evaluating the impact factor of journals and authors and allows for the recognition of new research issues. Different techniques for classifying citations have been proposed. Citations that provide background knowledge in the citing document have been classified as non-important or incidental by previous researchers. Citations that extend previous work in the citing document are classified as important. The accuracy achieved by existing citation models is not much higher. Better features need to be included for accurate predictions. A hybrid approach would present all possible combinations of cue-words and in-text citation-based features for citation classifications.

@article{Hussain2019,
  title     = {A Comprehensive Evaluation of Cue-Words based Features and In-text Citations based Features for Citation Classification},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {10},
  number    = {7},
  year      = {2019},
  publisher = {The Science and Information Organization},
  author    = {Syed Jawad Hussain and Sohail Maqsood and NZ Jhanjhi and Azeem Khan and Mahadevan Supramaniam and Usman Ahmed},
  doi       = {10.14569/IJACSA.2019.0100730},
  url       = {https://doi.org/10.14569/IJACSA.2019.0100730}
}