Web Information Extraction for Social Good: Food Pantry Answering As an Example. Chen, H. & Yu, H. In Austin, TX, May, 2023. ACM. The Web Conference 2023, Austin TX
doi  abstract   bibtex   
Social Determinants of Health (SDH) have more influence on health outcome than clinical care or the physical environment, namely food insecurity, housing instability, and health literacy. Many researchers design applications as a bridge to connect between resource providers and the deprived population. In this study, we take food pantries as a solution to mitigate food insecurity as an example to illustrate an automatic system combining location-aware information retrieval, web information extraction and domain-specific answering. To acquire the latest knowledge, our proposed framework first retrieves pantry candidates based on geolocation of the user, and utilizes structural information from markup language to extract semantic chunks related to six common requests. We use BERT and RoBERTa as information extraction models and compare three different web page segmentation methods in the experiments.
@inproceedings{chen_web_2023,
	address = {Austin, TX},
	title = {Web {Information} {Extraction} for {Social} {Good}: {Food} {Pantry} {Answering} {As} an {Example}},
	doi = {10.1145/3543507.3583880},
	abstract = {Social Determinants of Health (SDH) have more influence on health outcome than clinical care or the physical environment, namely food insecurity, housing instability, and health literacy. Many researchers
design applications as a bridge to connect between resource providers and the deprived population. In this study, we take food pantries as a solution to mitigate food insecurity as an example to illustrate an automatic system combining location-aware information retrieval, web information extraction and domain-specific
answering. To acquire the latest knowledge, our proposed framework first retrieves pantry candidates based on geolocation of the user, and utilizes structural information from markup language to
extract semantic chunks related to six common requests. We use BERT and RoBERTa as information extraction models and compare three different web page segmentation methods in the experiments.},
	publisher = {ACM},
	author = {Chen, Huan-Yuan and Yu, Hong},
	month = may,
	year = {2023},
	note = {The Web Conference 2023, Austin TX},
}

Downloads: 0