Pith. Sign in

RECORD

Training like Playing: A Reinforcement Learning And Knowledge Graph-based framework for building Automatic Consultation System in Medical Field

PaperPDF from arXiv

data files

Paper breakdownThe authors' claims, conditions and measurements, as JSON Signed receiptProof this review has not been altered
X Bluesky LinkedIn Reddit Hacker News
Yining Huang, Meilian Chen, Keke Tang

Machine Learning · Submitted 14 Jun 2021

abstract compiled · 2026-09-23 · no language-model review

The authors' main claim According to experiment we designed for evaluating its performance, it archives a good result.

how it is stated · asserted 0.98 read the record →

Comment (0) Read on arXiv ↗
Cite ▾
arXiv ID2106.07502 (version 1)
Pith IDUGWV7XKE, Pith's permanent record of this paper and its review
CitationKeke Tang, Meilian Chen, Yining Huang. "Training like Playing: A Reinforcement Learning And Knowledge Graph-based framework for building Automatic Consultation System in Medical Field." arXiv:2106.07502 (2021). https://pith.science/paper/UGWV7XKE
Claim paper

The compiled record

The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Open the compiled record

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 3,527 input tokens.

What the abstract itself states, as probabilities: claims first 0.04 · states limitation 0.15 · human participants 0.67 · controlled comparison 0.05 · randomized assignment 0.02 · states quantitative result 0.04 · compares with prior results 0.07 · mentions code or data availability 0.05.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; .

We introduce a framework for AI-based medical consultation system with knowledge graph embedding and reinforcement learning components and its implement. Our implement of this framework leverages knowledge organized as a graph to have diagnosis according to evidence collected from patients recurrently and dynamically. According to experiment we designed for evaluating its performance, it archives a good result. More importantly, for getting better performance, researchers can implement it on this framework based on their innovative ideas, well designed experiments and even clinical trials.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#sentencerolethe authors' own workstates the main conclusionstrength
s0 We introduce a framework for AI-based medical consultation system with knowledge graph embedding and reinforcement learning components and its implement. contribution 1.00 0.89 0.11 asserted 0.99
s1 Our implement of this framework leverages knowledge organized as a graph to have diagnosis according to evidence collected from patients recurrently and dynamically. method 0.98 0.75 0.09 asserted 1.00
s2 According to experiment we designed for evaluating its performance, it archives a good result. result 1.00 0.94 0.57 asserted 0.98
s3 More importantly, for getting better performance, researchers can implement it on this framework based on their innovative ideas, well designed experiments and even clinical trials. future work 1.00 0.07 0.17 hedged possibility 0.77

What kind of paper the classifier reports this is: method or tool 1.00 · evidence from observation or experiment 0.96.

What was examined, and what was not

sources read
  • arXiv listing abstract v1
not examined
  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers
levels run
  • P abstract
not run
  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

The paper
Original abstract

We introduce a framework for AI-based medical consultation system with knowledge graph embedding and reinforcement learning components and its implement. Our implement of this framework leverages knowledge organized as a graph to have diagnosis according to evidence collected from patients recurrently and dynamically. According to experiment we designed for evaluating its performance, it archives a good result. More importantly, for getting better performance, researchers can implement it on this framework based on their innovative ideas, well designed experiments and even clinical trials.

Human review

How Pith checks papers → · Full audit of this paper →

How the literature uses it

What this paper does with the arXiv papers it cites

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

What it cites · 10 sentences without a confident label

Uses the method or tool of 2 papers
  • It’s inspired by the dropout regularization method proposed in [ref] and leveraged in [ref] for introducing noise to make model more generalized.

    Improving neural networks by preventing co-adaptation of feature detectors · Diagnosis Network · 0.93

  • Finally, in order to increase the randomness in the training process and encourage the actor to find a better path of consultation from a more diversified perspective, we use Boltzmann Exploration [ref] method.

    Boltzmann Exploration Done Right · Parameter Setting and Training Process · 1.00

Cites as background 1 paper

Discussion

Know something about this paper? Leave a comment, or highlight any passage above to comment on it.