A scalable search system over e-commerce databases

大多数高质量的电子商务数据（如Shopify）都是结构化的，或者至少是半结构化的。结构化数据通过编写查询（如SQL）或通过预定义的搜索表单提供。编写 **SQL查询需要理解通常很复杂的数据库模式，仅限于技术用户，要修剪 ** 错误，而且很耗时。使用预定义的搜索表单仅限于表单集。如果需要一个新的表单，它需要由开发人员设计，这又是耗时和昂贵的。在这个建议中，** 我们使用关键字搜索范式。关键字搜索范式的主要优点是它的易用性 ** 和非技术用户的熟悉度。如果有人曾经使用过谷歌（或其他网络搜索引擎），她可以 ** 立即使用我们的系统没有培训。此外，一组关键字提供了我们找到相关答案所需的表达水平。Warranty Life通过集成到电子商务系统中，为各种产品提供不同的保修。为了深入了解客户的行为和购物模式，Warranty Life需要探索 ** 电子商务数据库，以找到不同实体之间的关系。电子商务 ** 数据库上的关键字搜索系统是满足这一需求的完美解决方案。它使Warranty Life在市场上具有竞争优势，并 ** 使其能够吸引更多的客户。**在本研究中，我们的目标是建立一个新的保修期搜索系统。该搜索系统将 ** 增加Warranty Life的收入，并将直接有利于加拿大，因为Warranty**Life的更多收入意味着该公司在加拿大的更多投资，并雇用更多的加拿大人才，以扩大现有的 ** 服务和产品与开发的技术。

Most of high-quality e-commerce data (such as Shopify) are structured or at least semi-structured. Accessing**structured data are provided by writing queries (such as SQL), or through pre-defined search forms. Writing**SQL queries needs the understanding of often complex database schema, is limited to technical users, is prune**to error, and is time consuming. Using pre-defined search forms are limited to the set of forms. If a new form is**required, it needs to be designed by developers which is again time consuming and expensive. In this proposal,**we use the keyword search paradigm. The main advantage of the keyword search paradigm is its ease of use**and familiarity for non-technical users. If anyone ever used Google (or other Web search engines) she can**instantly use our system with no training. Furthermore, a set of keywords provide the level of expressiveness**that we need to find relevant answers.**Warranty Life provides different warranties to a variety of products by integrating into e-commerce systems. In**order to gain insight into customer's behaviors and shopping patterns, Warranty Life needs to explore**e-commerce databases to find relations between different entities. A keyword search system over e-commerce**databases is the perfect solution to this need. It gives Warranty Life a competitive advantage in the market and**enable it to attract more customers.**In the proposed research, our goal is to build a new search system for Warranty Life. The search system will**increase the revenue of Warranty Life and will be directly beneficial to Canada as more revenue for Warranty**Life means more investment from the company in Canada and hiring more Canadian talent to expand existing**offers and products with respect to the developed technology.