arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1808.09479cs.CL
收藏

用于社区社交媒体预测任务的剩余因子自适应

Residualized Factor Adaptation for Community Social Media Prediction Tasks

Stony Brook University(石溪大学);University of Pennsylvania(宾夕法尼亚大学)

Mohammadzaman Zamani, H. Andrew Schwartz, Veronica E. Lynn, Salvatore Giorgi, Niranjan Balasubramanian

摘要:

社交媒体语言的预测模型在捕捉社区成果方面表现出了希望,但迄今为止的方法在很大程度上忽视了语言起源社区的社会人口背景(例如年龄,教育率,种族)。例如,假设阿拉巴马州莫比尔的人口年龄相对较大,使用单词的方式与旧金山的人相同可能是不准确的,旧金山的中位年龄更年轻,大学教育率更高。在本文中,我们提出了剩余因子适应,社区预测任务的一种新方法,既(a)有效地整合社区属性,以及(b)适应语言特征的社区属性(因素)。我们使用11个人口统计学和社会经济学属性,并在五个不同的社区层面预测任务中评估我们的方法,包括健康(心脏病死亡率,健康状况良好/不佳的百分比),心理学(生活满意度)和经济学(房价上涨百分比,止赎率)。我们的评估表明,剩余因素适应显着提高了4的5个社区水平的结果预测比以前的国家的最先进的纳入社会人口背景。

英文摘要:

Predictive models over social media language have shown promise in capturing community outcomes, but approaches thus far largely neglect the socio-demographic context (e.g. age, education rates, race) of the community from which the language originates. For example, it may be inaccurate to assume people in Mobile, Alabama, where the population is relatively older, will use words the same way as those from San Francisco, where the median age is younger with a higher rate of college education. In this paper, we present residualized factor adaptation, a novel approach to community prediction tasks which both (a) effectively integrates community attributes, as well as (b) adapts linguistic features to community attributes (factors). We use eleven demographic and socioeconomic attributes, and evaluate our approach over five different community-level predictive tasks, spanning health (heart disease mortality, percent fair/poor health), psychology (life satisfaction), and economics (percent housing price increase, foreclosure rate). Our evaluation shows that residualized factor adaptation significantly improves 4 out of 5 community-level outcome predictions over prior state-of-the-art for incorporating socio-demographic contexts.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑