At Kai Analytics, our mission is to help our clients leverage their data to produce unbiased insights, naturally and effectively. To this end, we’ve been developing Unigrams, a qualitative data analysis platform that makes the tools we use when consulting accessible to the everyday user. Unigrams helps analysts leverage Natural Language Processing technology and was originally designed to help higher education administrators quickly understand the needs of their students and faculty. We encourage you to check it out and try the free previewer by clicking the image below.
In our first Lightning Talk, we wanted to show off some of the basic features that make Unigrams unique.
The first thing you may notice when uploading your dataset to Unigrams is the option to choose where you want your data stored. At Kai Analytics, we understand how important it is to guarantee to stakeholders that their data is safe. Unigrams does this by giving you the option to store data on servers located in either the US or Canada. Located somewhere else? No problem, just reach out and we can set up servers in a country of your choice.
Since Unigrams is built for qualitative analysis, this update focused on the Text Analysis dashboard. On the left, an interactive network graph visually displays themes and connections. To the right is the topic modelling display, with an Inter-topic distance map showing all themes, and a list of the Top 30 most salient terms for the selected theme. For a more detailed breakdown, a chart displays all analyzed comments, along with the associated themes and probabilities. This chart can be downloaded in the top right corner, to keep Unigrams from becoming a Blackbox.
A major promise of most qualitative studies is the guarantee of anonymity. Unigrams enables this by automatically editing out sensitive information, such as names and phone numbers.
Our Lightning talks are a great way to stay up to date on new features and have a say in the development process! To hear about the next one, make sure you subscribe to our newsletter (form at the bottom of the page) or follow us on LinkedIn.
Comments