Big data processing is an urgent and unresolved challenge that originates from the intensive development of information technology. The recent techniques lose their effectiveness rapidly as the volumes of data increase. In this article, we will put down our vision of the basic approaches and models related to problem solving, based on processing large data volumes. This article introduces a two-stage decomposition of a problem, related to assessing management options. The first stage of our original approach implies a semantic analysis of textual information; the second stage is built around finding association rules in a database, processing them via mathematical statistics methods, and converting data and objectives to a vector. We suggest processing the collected news events by a semantic model, which describes their key features and interconnections between them in a specified subject area. The classification-based association rules allow assessing the likelihood of a particular event using a set chain of events. This approach can be applied through the analysis of online news in a specified market segment.

大数据处理是迫切且尚未解决的挑战，它源于信息技术的密集发展。随着数据量的增加，最近的技术迅速失去了有效性。在本文中，我们将基于处理大数据量，提出与问题解决相关的基本方法和模型的愿景。本文介绍了与评估管理选项相关的问题的两阶段分解。我们原始方法的第一阶段意味着对文本信息进行语义分析。第二阶段围绕查找数据库中的关联规则，通过数学统计方法进行处理以及将数据和目标转换为向量而构建。我们建议通过语义模型处理收集到的新闻事件，描述了它们的关键功能以及它们在指定主题领域之间的相互联系。基于分类的关联规则允许使用一组事件链来评估特定事件的可能性。可以通过分析指定市场细分中的在线新闻来应用此方法。