SBIR Phase I: Intelligent Word Completion

这个小型企业创新研究第一阶段的研究项目调查了一个新的概念和方法，以引入基于自然语言处理领域的想法的智能单词补全软件，并测试其在多种语言上的可行性。这项研究的结果应该使软件能够更准确地预测用户的意图。 广义地说，数据输入是所有使用电子设备的人的一项普遍任务，这些电子设备包括计算机、手持设备（如手机和个人数字助理）以及身体残疾者专用设备。由于广泛意义上的各种数据的输入非常普遍，因此在这一研究项目下开发的软件将通过提高普通用户的生产力而为他们增加很大的价值。这一软件还可能使残疾人受益，他们对这类软件的需求甚至比普通人更大，因为他们的残疾使数据输入这一现代社会中每个人都需要的一项要求加倍困难。它还可以帮助其他弱势群体，他们可能没有完全识字，可能无法正确拼写许多单词，但可能需要创建各种文件。除此之外，它的使用将增加我们对统计智能在各种自然语言处理任务中的作用的科学理解，以及如何使用它来提高自然语言软件的性能。

This Small Business Innovation Research Phase I research project investigates a novel concept and method to introduce intelligent word completion software based on ideas from the field of natural language processing and test its feasibility over multiple languages. This outcome of this research should enable the software to more accurately predict the intent of the user. Data entry, broadly interpreted, is a ubiquitous task for all persons using electronic devices such as computers, handheld devices such as cell phones and personal digital assistants, and special devices for the physically handicapped. The software developed under this research project will add much value to ordinary users by increasing their productivity since the entry of data of all kinds, broadly construed, is very widespread. This software can also potentially benefit disabled persons whose need for such software is even greater than that of ordinary persons, as their disability makes data entry, a requirement for everyone in modern society, doubly difficult. It can also help other disadvantaged groups who may not be fully literate and may not be able to spell many words correctly and may yet have a need to create documents of various kinds. Beyond this, its use will add to our scientific understanding of the role of statistical intelligence in various natural language processing tasks and how it may be used to enhance performance of natural language software.