This paper describes the graphical user interface design and implementation of LexOnt, a semi-automatic ontology creation tool. LexOnt is built as a Protégé plugin and uses the Programmable Web directory as its corpus. Current semi-automatic ontology creation systems are domain specific and rely on already structured text. LexOnt, on the other hand, is built specifically for those who are not experts within a domain, but want to understand the domain on a high-level and create an ontology that describes it. Using well-known NLP algorithms, LexOnt generates a list of top terms and phrases from the Programmable Web corpus to enable users to find high-level features that distinguish one Programmable Web service category from another. To also aid non-experts in a domain, LexOnt relies on outside sources such as Wikipedia and Wordnet to help the user identify the important terms within a service category. We describe the details of LexOnt's interactive interface that allows users to semi-automatically find top terms and phrases of a service domain and create ontology entities describing it with these features.