Online Doctor Recommendation with Convolutional Neural Network and Sparse Inputs

Abstract

The recommendation system in the online medical consultation website is a system to assist patients to find appropriate doctors. Based on the analysis of the current situation of the development of an online medical community (Haodf.com) in China, this paper puts forward recommendation suggestions of finding the right hospital and doctor to promote the rapid integration of Internet technology and traditional medical services. A new recommendation model called Probabilistic Matrix Factorization integrated with Convolutional Neural Network (PMF-CNN) is proposed in the paper. Doctors’ data in Haodf.com were used to evaluate the performance of our system. The model improves the performance of medical consultation recommendations by fusing review text and doctor information based on CNN (Convolutional Neural Network). Specifically, CNN is used to learn the feature representation of the review text and the doctors’ information. Furthermore, the extended matrix factorization model is exploited to fuse the review information feature and the initial value of the doctors’ information for recommendation. As is shown in the experimental results on Haodf.com datasets, the proposed PMF-CNN achieves better recommendation performances than the other state-of-the-art recommendation algorithms. And the recommendation system in an online medical website improves the utilization efficiency of doctors and the balance of public health resources allocation.

Machine and cognitive intelligence for human health: systematic review

“…Third, Sahu et al [ 90 ] mentioned collecting “demographic information from a subset of existing ecobee users to understand the association between age, sex, and other relevant demographic indicators (p. 8)”, and Arsevska et al [ 79 ] indicated integrating geographical and language factors. Other directions include: (1) integrating additional physiological signal monitoring modules [ 91 ], (2) combining the temporality of messages in clustering [ 61 ], (3) adopting linked open data as complementary answer sources [ 30 ], (4) exploiting health data produced via passive smartphone sensing technologies and linking them with Web-based applications [ 92 ], (5) integrating additional types of mappings or services with the basis of clinical guidelines to allow linking electronic health records with guideline-oriented decision support applications [ 93 ], (6) integrating multiple context information based on deep learning [ 94 ], (7) allowing seamless integration of data from varied sources or repositories [ 54 ], and (8) collecting propagation-related information and time series information to enhance model performance [ 86 ].…”

A deep learning model for online doctor rating prediction

“…We benchmark ODRP‐HABiLSTM with four state‐of‐the‐art ODRP models: MF (Zhang et al, 2017), iDoctor (Zhang et al, 2017), PMF‐CNN (Yan et al, 2020), and RR&R‐CNN (Zhou et al, 2021). Grid search is used to search the hyperparameters of the benchmark models.…”

“…Online doctor ratings are considerably affected by patient preferences and doctor attributes (Zhang et al, 2017). Auxiliary information, such as textual reviews, capture this contextual information and have been used to support the information provided by the patient–doctor rating matrix and enhance the accuracy of recommendations (Yan et al, 2020; Zhang et al, 2017; Zhou et al, 2021). In addition, textual reviews give additional insights into the doctors' attributes (Lin et al, 2020) and preferences of each patient (Janssen & Lagro‐Janssen, 2012), which are not captured in the numerical ratings.…”

Music Recommendation System and Recommendation Model Based on Convolutional Neural Network

“…They preprocess the input data to 0 or 1 based on the convolutional neural network, which can fully save the data storage and space, and the side help improves the efficiency of the recommendation. They also use RBF to establish a kinship network and make recommendations based on useful information screened out by similar users in the kinship network [8]. However, the above-mentioned studies need to learn a large number of parameters, which require a large amount of labeled training data and a large amount of professional knowledge to ensure correctness, ignoring issues such as the importance of the strength of the relationship between users of different levels in the network.…”