@article{fdi:010079377,
  title = {{S}emi-supervised text classification framework : an overview of dengue landscape factors and satellite earth observation},
  author = {{L}i, {Z}. {C}. and {G}urgel, {H}. and {D}essay, {N}adine and {H}u, {L}. {J}. and {X}u, {L}. and {G}ong, {P}.},
  editor = {},
  language = {{ENG}},
  abstract = {{I}n recent years there has been an increasing use of satellite {E}arth observation ({EO}) data in dengue research, in particular the identification of landscape factors affecting dengue transmission. {S}ummarizing landscape factors and satellite {EO} data sources, and making the information public are helpful for guiding future research and improving health decision-making. {I}n this case, a review of the literature would appear to be an appropriate tool. {H}owever, this is not an easy-to-use tool. {T}he review process mainly includes defining the topic, searching, screening at both title/abstract and full-text levels and data extraction that needs consistent knowledge from experts and is time-consuming and labor intensive. {I}n this context, this study integrates the review process, text scoring, active learning ({AL}) mechanism, and bidirectional long short-term memory ({B}i{LSTM}) networks, and proposes a semi-supervised text classification framework that enables the efficient and accurate selection of the relevant articles. {S}pecifically, text scoring and {B}i{LSTM}-based active learning were used to replace the title/abstract screening and full-text screening, respectively, which greatly reduces the human workload. {I}n this study, 101 relevant articles were selected from 4 bibliographic databases, and a catalogue of essential dengue landscape factors was identified and divided into four categories: land use ({LU}), land cover ({LC}), topography and continuous land surface features. {M}oreover, various satellite {EO} sensors and products used for identifying landscape factors were tabulated. {F}inally, possible future directions of applying satellite {EO} data in dengue research in terms of landscape patterns, satellite sensors and deep learning were proposed. {T}he proposed semi-supervised text classification framework was successfully applied in research evidence synthesis that could be easily applied to other topics, particularly in an interdisciplinary context.},
  keywords = {dengue ; landscape ; satellite {E}arth observation ; deep active learning ; natural language processing},
  booktitle = {},
  journal = {{I}nternational {J}ournal of {E}nvironmental {R}esearch and {P}ublic {H}ealth},
  volume = {17},
  numero = {12},
  pages = {art. 4509 [29 ]},
  year = {2020},
  DOI = {10.3390/ijerph17124509},
  URL = {https://www.documentation.ird.fr/hor/fdi:010079377},
}