講座題目：Short-Form Videos and Mental Health: A Knowledge-Guided Neural Topic Model（短視頻與心理健康：一種知識(shí)引導(dǎo)的神經(jīng)主題模型）
主講人：柴一棟 香港城市大學(xué)
講座時(shí)間：2025年12月19日10:30
講座地點(diǎn)：學(xué)院216
內(nèi)容摘要：
Along with the rise of short-form videos, their mental impacts on viewers have led to widespread consequences, prompting platforms to predict videos' impact on viewers' mental health. Subsequently, they can take intervention measures according to their community guidelines. Nevertheless, applicable predictive methods lack relevance to well-established medical knowledge, which outlines clinically proven external and environmental factors of mental disorders. To account for such medical knowledge, we resort to an emergent methodological discipline, seeded Neural Topic Models (NTMs). However, existing seeded NTMs suffer from the limitations of single-origin topics, unknown topic sources, unclear seed supervision, and suboptimal convergence. To address those challenges, we develop a novel Knowledge-Guided NTM to predict a short-form video's suicidal thought impact on viewers. Extensive empirical analyses using short-form videos such as Douyin prove that our method outperforms state-of-the-art benchmarks. Our method also discovers medically relevant topics from videos that are linked to suicidal thought impact. We contribute to IS with a novel video analytics method that is generalizable to other video classification problems. Practically, our method can help platforms understand videos' suicidal thought impacts, thus moderating videos that violate their community guidelines.
隨著短視頻的興起，其對(duì)觀眾心理健康的影響引發(fā)廣泛社會(huì)關(guān)注，促使各大平臺(tái)開(kāi)始預(yù)測(cè)視頻內(nèi)容對(duì)觀眾心理健康的潛在影響，并根據(jù)社區(qū)準(zhǔn)則采取干預(yù)措施。然而，現(xiàn)有的預(yù)測(cè)方法往往缺乏與臨床醫(yī)學(xué)知識(shí)的關(guān)聯(lián)，而這些知識(shí)系統(tǒng)闡述經(jīng)臨床驗(yàn)證的精神障礙外在與環(huán)境誘因。為融入醫(yī)學(xué)專(zhuān)業(yè)知識(shí)，團(tuán)隊(duì)引入一種新興方法學(xué)分支——種子神經(jīng)主題模型。但現(xiàn)有種子神經(jīng)主題模型存在主題來(lái)源單一、主題溯源困難、種子監(jiān)督機(jī)制不明確及收斂效果欠佳等局限性。為克服這些挑戰(zhàn)，團(tuán)隊(duì)開(kāi)發(fā)了一種新型知識(shí)引導(dǎo)神經(jīng)主題模型，專(zhuān)門(mén)用于預(yù)測(cè)短視頻內(nèi)容對(duì)觀眾意念的影響。通過(guò)對(duì)抖音等平臺(tái)短視頻的大規(guī)模實(shí)證分析表明，該方法在預(yù)測(cè)性能上超越當(dāng)前最先進(jìn)的基準(zhǔn)模型。同時(shí)，該方法能從視頻內(nèi)容中識(shí)別出某些醫(yī)學(xué)主題，為信息系統(tǒng)領(lǐng)域貢獻(xiàn)了可推廣至其他視頻分類(lèi)問(wèn)題的創(chuàng)新分析方法。在實(shí)踐中，該方法能幫助平臺(tái)有效識(shí)別具有心理誘導(dǎo)風(fēng)險(xiǎn)的視頻內(nèi)容，從而對(duì)違反社區(qū)準(zhǔn)則的視頻進(jìn)行及時(shí)管控。
主講人簡(jiǎn)介：
柴一棟，香港城市大學(xué)長(zhǎng)聘副教授，合肥工業(yè)大學(xué)博士生導(dǎo)師，國(guó)家高層次青年人才。博士畢業(yè)于清華大學(xué)經(jīng)管學(xué)院管理科學(xué)與工程系（信息系統(tǒng)方向），主要關(guān)注如何設(shè)計(jì)創(chuàng)新性的人工智能方法，更好地服務(wù)于個(gè)人、組織和社會(huì)的現(xiàn)代科學(xué)化管理。以第一作者或通訊作者發(fā)表研究成果于MISQ、ISR、JMIS等管理信息系統(tǒng)頂刊（UTD/FT），IEEE TDSC、IEEE TPAMI、IEEE TKDE、ACM TOIS等信息安全/人工智能/數(shù)據(jù)挖掘等領(lǐng)域的期刊（CCF A），以及《管理科學(xué)學(xué)報(bào)》等中文期刊。擔(dān)任ACM Transactions on AI Security and Privacy、《系統(tǒng)科學(xué)與系統(tǒng)工程學(xué)報(bào)》（英文版）、Industrial Management & Data Systems等期刊副主編。
