Home
Implicit Signal Mining for Personalized Decision Systems

Chika Komari

A unified track across structural, semantic, and psychological signal mining

I focus on implicit signal mining in complex information networks for risk-aware and personalized decision systems.

O1-KR1 (due June 30, 2026): critical node identification on multi-layer heterogeneous networks O1-KR2 (due September 30, 2026): scalable financial knowledge-graph infrastructure and signal mining automation O2-KR1 (due December 31, 2026): personalized multi-device AI companion based on Airi

Publications / Preprint

3

1 under review · 1 accepted · 1 NeurIPS poster preprint

GPA / Rank

4.04/5.00 · 15/154

Sun Yat-sen University (School of Intelligent Systems Engineering)

Kaggle Models Views / Downloads

2.7K+ / 1.0K+

CSIRO Biomass Prediction Model Series

Kaggle Tier

Notebooks Expert

Kaggle Progression tier (2025-Present)

Internship

Evolution Asset Management

AI Intern (Apr 2025 - Present)

GitHub Contributions

800+

Public contribution activity across projects

GitHub Commit Activity

Open GitHub

Last 12 months

Syncing GitHub profile contributions...

Source: GitHub profile graph

Live contributions are fetched from your public GitHub profile graph.

Roadmap Snapshot

View Full Roadmap

Psychological Signal Mining

Avg. progress 11.1% · 1 roots / 3 nodes
Planned Psychological Signal Mining (Personalized Decision and Interaction)

Semantic Signal Mining

Avg. progress 11.1% · 1 roots / 3 nodes
Ongoing Semantic Signal Mining (Financial Knowledge Graphs)

Structural Signal Mining

Avg. progress 11.1% · 1 roots / 3 nodes
Ongoing Structural Signal Mining (Critical Nodes in Complex Networks)

Featured Artifacts

Browse All Artifacts
Alignment Between the Minds (ICML 2026 Submission)

Alignment Between the Minds (ICML 2026 Submission)

研究具身体在多轮指令交互中对用户潜在偏好的持续推断、维护与利用，聚焦高维心理信号挖掘。

paper Latent Preference DiscoveryEmbodied Agent
Identifying Critical Nodes with Deep Learning and Reinforcement Learning

Identifying Critical Nodes with Deep Learning and Reinforcement Learning

研究城市路网场景下的关键节点识别与影响力传播建模，属于低维结构信号挖掘方向的核心成果。

paper Complex NetworkCritical Node Identification
AIRI LocalGUI Task Agent (Open Source; In Progress)

AIRI LocalGUI Task Agent (Open Source; In Progress)

把自然语言指令转换为可执行桌面任务计划，支持自动校验、自动修复与多工具协同执行。

project Desktop AutomationPlanner-Executor

Latest Updates

Open Blog
Google 软件工程-风格指南与规则

Google 软件工程-风格指南与规则

在进行软件开发的流程中，设置和持续维护一套风格指南与规则，用于指导开发至关重要。

2026/03/18 software_engineering
Google 软件工程-团队领导与生产力度量

Google 软件工程-团队领导与生产力度量

在进行多人协作的过程中，需要发挥良好的团队领导才能并合理度量生产力，尽可能地用合理的方式解决问题。

2026/03/11 software_engineering
Google 软件工程-团队合作与知识管理

Google 软件工程-团队合作与知识管理

在进行项目时，如何更好地参与团队合作与进行知识管理，是一个值得思考的问题。

2026/03/04 software_engineering
Google 软件工程-序言

Google 软件工程-序言

软件工程重点考虑时间、规模和权衡。书中提出的海勒姆定律和碧昂斯规则都是值得反复思考和品味的重要原则。书中提出的经典案例也引人深思。

2026/03/02 software_engineering
讨论：从《DeepSeek-R1 通过强化学习激励LLM中的推理》开始

讨论：从《DeepSeek-R1 通过强化学习激励LLM中的推理》开始

关键观点 ：LLM 的推理能力可以通过纯强化学习 (RL) 来激励，从而无需人工标记的推理轨迹

2025/09/23 llm_learning
  • Jan 2026: submitted first-author ICML 2026 paper on latent preference discovery
  • 2025: accepted at GBCESC 2025 and received Best Paper Award
  • Since Apr 2025: building financial knowledge-graph infrastructure at Evolution Asset Management