I-Corps: Social Media Misinformation Interactive Dashboard

这个I-Corps项目更广泛的影响/商业潜力是开发一个在线仪表板，其中包含错误信息预测趋势和分析，以帮助解决美国流行的错误信息。错误信息是一个重大问题，可能会造成对基本主题的混淆和误解。网上的错误信息可能会导致人们质疑循证医学指导或拒绝安全治疗。了解当前的错误信息内容和趋势有助于企业实体和社交媒体用户。对于企业来说，所提出的工具套件可以通过探索和预测有关错误信息主题的公众情绪来显示错误信息如何影响企业。这种分析可以通过将资源从人工识别转移到分析错误信息帖子的内容和影响来为这些机构提供价值。个人用户可能会更好地装备，不仅了解错误信息的主要主题，而且还解释了为什么这个错误信息的主题正在传播。潜在的客户可能会被吸引到这项拟议的技术，因为信息需求是从基层组织，这可能是不一致的数据质量。这个I-Corps项目基于自动数据收集，数据分析和深度学习方法的开发。目标是开发一个应用程序和相关网站，集中最新的错误信息内容和指标。 这样的解决方案将为潜在客户提供一个企业级系统，以帮助更好地了解社交媒体平台上传播的错误信息的影响和类型。目前，正在开发一系列与错误信息内容互动的工具。正在构建管道，以便使用selenium框架为收集步骤传输和存储原始数据。这些数据存储在本地，并将作为原始信息来支持最终的仪表板。已经开发了经过训练以识别基于文本的错误信息的深度学习模型，其目标是扩展到图像，声音和视频识别，以应对当前的社交内容趋势。该变形金刚框架还识别对指定主题组的情绪以支持分析。该技术涉及大数据、自然语言处理、人工智能和统计分析等多个研究领域，体现了NSF的法定使命，通过基金会的智力价值和更广泛的影响力评审标准进行评估，被认为值得支持。

The broader impact/commercial potential of this I-Corps project is the development of an online dashboard with misinformation forecast trends and analysis to help address the misinformation endemic in America. Misinformation is a significant issue that may create confusion and misunderstanding about essential topics. Misinformation online may result in people questioning evidence-based medical guidance or refusing safe treatments. Understanding current misinformation content and trends supports both corporate entities and social media users. For corporations, th proposed suite of tools may show how misinformation impacts businesses by exploring and forecasting public sentiment concerning relevant misinformation topics. This analysis may provide value to these agencies by shifting resources from manual identification to analyzing the content and implications of misinformation posts. Individual users may be better equipped by understanding not only the major topics in misinformation but also an explanation of why this misinformation topic is being spread. Potential customers may be drawn to this proposed technology since information demand is met from grassroots organizations, which can be inconsistent with data quality. This I-Corps project is based on the development of automated data collection, data analytics, and deep learning methodologies. The goal is to develop an application and associated website that centralizes up-to-date misinformation content and metrics. Such a solution will provide potential customers with an enterprise-level system to help better understand the implications and types of misinformation spread across social media platforms. Currently, an array of tools to interact with misinformation content is under development. Pipelines are being constructed to channel and store raw data using the selenium framework for the collection step. Thesedata are stored locally and will act as the raw information to support the finalized dashboard. Deep learning models that are trained to identify text-based misinformation have been developed with the goal of expanding to image, sound, and video identification to address current social content trends. This Transformers framework also identifies sentiment on specified topic groups to support analysis. The proposed technology involves multiple research areas, including big data, natural language processing, artificial intelligence, and statistical analysis.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.