Educational MCP Server

"""知识追踪模块（规则版占位），可替换为 DKVMN/AKT。""" from __future__ import annotations from collections import defaultdict from typing import Dict, List import database from schemas import KnowledgeTracingRequest, KnowledgeTracingResponse, SkillInteraction, SkillProgress def _update_probability(prob: float, interaction: SkillInteraction) -> float: prob = max(min(prob, 0.999), 0.001) learning_rate = 0.3 forgetting = 0.4 if interaction.correct: prob = prob + (1 - prob) * learning_rate else: prob = prob * forgetting if interaction.confidence is not None: prob = 0.7 * prob + 0.3 * interaction.confidence if interaction.time_spent_seconds and interaction.time_spent_seconds > 120: prob -= 0.05 return max(min(prob, 1.0), 0.0) def trace(payload: KnowledgeTracingRequest) -> KnowledgeTracingResponse: prob_map: Dict[str, float] = defaultdict(lambda: 0.5) prob_map.update(payload.prior_mastery) history: Dict[str, List[bool]] = defaultdict(list) for interaction in payload.interactions: prev = prob_map[interaction.skill] updated = _update_probability(prev, interaction) prob_map[interaction.skill] = updated history[interaction.skill].append(interaction.correct) skills: List[SkillProgress] = [] for skill, probability in prob_map.items(): recent = history.get(skill, []) if len(recent) >= 2 and recent[-2:] == [True, True]: trend = "上升" elif recent and not recent[-1]: trend = "下降" else: trend = "平稳" if probability >= 0.85: action = "安排挑战题巩固迁移。" elif probability >= 0.6: action = "保持混合题训练，关注错误类型。" else: action = "回到基础例题，配合讲解反馈。" database.set_mastery(payload.student_id, skill, probability) skills.append( SkillProgress( skill=skill, probability_mastery=round(probability, 3), trend=trend, next_action=action, ) ) recommended_sequence = [item.skill for item in sorted(skills, key=lambda s: s.probability_mastery)] return KnowledgeTracingResponse( request_id=payload.request_id, student_id=payload.student_id, skills=skills, recommended_sequence=recommended_sequence, model_version="rule-0.1", )