The HoPE Model Architecture: a Novel Approach to Pregnancy Information Retrieval Based on Conversational Agents

Abstract

Conversational agents are used to communicating with humans in a friendly manner. To achieve the highest level of performance, agents need to respond assertively and fastly. Transformer architectures are shown to produce excellent performances on recent tasks; however, for tasks involving conversational agents, they may have a lower speed performance. The main goal of this study is to evaluate and propose a HoPE (Healthcare Obstetric in PrEgnancy) model that is tailored to pregnancy data. We carried out a dataset extraction and construction process based on collections of health documents related to breastfeeding, childcare, pregnant care, nutrition, risks, vaccines, exams, and physical exercises. We evaluated two pre-trained models in the Portuguese language for the conversational agent architecture proposal and chose the one with the best performance to compose the HoPE architecture. The BERTimbau model, which has been trained on data augmentation strategies, proves to be able to retrieve information quickly and most accurately than others. For the fine-tuning process, we achieved a Spearman correlation of 95.55 on BERTimbau augmented with a few pairs (1.500 pairs). The HoPE model architecture achieved an F1-Score of 0.89, outperforming other combinations tested in this study. We will evaluate this approach for clinical studies in future studies.

Transparent, Low Resource, and Context-Aware Information Retrieval From a Closed Domain Knowledge Base

“…The authors discuss automatic subjective tag extraction from user utterances and online reviews using state-of-the-art machine learning techniques such as BERT, adversarial training, and data programming. HoPE [40] is an architecture for conversational agents that uses in its structure ontology-based modeling and Sentence-BERT [41] networks adjusted on pregnancy guidelines data to support pregnant women in obtaining more reliable information during the baby's thousand days period.…”

Productivity of incident management with conversational bots-a review

“…Similarly, according to, [39], [46], [61] there is a segment of users who use these bots in Processes for personal assistance for various reasons. It was also found in [20], [28], [32], [33], [36], [38], [58], [59], [63], that chatbots are used in customer service processes. In this analysis, the application of chatbots in incident management processes has not been identified; however, it has been mentioned that the teaching process (student-delivered course), customer service process (customer-supplier), health process (patient-doctor), processes for personal assistance (assisted-assistant) and even the training of bots (trained-trainer), are carried out by conversational bots; in the case of the Incident Management Process (user-technician), its application is highly necessary and important to increase productivity in this process.…”

“…In this analysis, the application of chatbots in incident management processes has not been identified; however, it has been mentioned that the teaching process (student-delivered course), customer service process (customer-supplier), health process (patient-doctor), processes for personal assistance (assisted-assistant) and even the training of bots (trained-trainer), are carried out by conversational bots; in the case of the Incident Management Process (user-technician), its application is highly necessary and important to increase productivity in this process. [37], [46]- [48], [53], [54], [59], [65] 3 Customer service processes 9 [23], [31], [35], [36], [39], [41], [61], [62], [66] 4 Health processes 11 [24], [25], [28], [29], [32]- [34], [43], [44], [56], [57] 5 Processes for personal assistance 3 [42], [49], […”