Getting started with Explorer
Gavagai Explorer is a web-based application that is best experienced using Google Chrome. You can download Chrome here.
Creating an account
You need to create an account with the Gavagai Explorer before you can start to analyse your texts. To get a free trial account, please contact us at sales@gavagai.io.
Creating a project
Gavagai Explorer allows you to retrieve insights and statistics from data. The first step in this process is creating a project from your data file or from an online source.
Getting your data ready
Preparing an upload file
Both file types that are compatible with the Explorer (.xlsx and .csv) store data in columns and rows. For the Explorer to correctly process your file, the uploaded data should be laid out in a particular way.
Each line of the file should correspond to one verbatim. According to your use-case, this could be a document, a product review or a tweet. All texts for analysis should be located in the same column of your file, as the explore process is carried out on a single column. We will refer to this as the 'main text column'. In other columns of the sheet, you can optionally include metadata about the text (e.g. date, location, author_id), which can later be used for reference or for filtering your data inside the Explorer.
The file should also begin with a header row, containing appropriate titles for each column.
Here you can find an example file containing hotel reviews that is laid out in the correct way. There are four columns containing metadata, plus a main text column which is entitled 'review'.
In .xlsx files, it is possible to leverage formulas for processing data, for example: to concatenate text or to calculate time intervals. This is completely compatible with the Explorer for both metadata and textual data. In pre-processing, the system knows to take the resulting value of the formula, rather than the text of the formula itself.
Getting online reviews
In some cases you may not have the data necessary for analysis. Explorer provides a way for you to get the data, mostly in the form of reviews, from a number of online sources, without any extra charge.
To get some data, go to the scrape data page and add either a url, a Google place id or a search query to start the scraping, as shown in the modal below:
On the resulting 'Get Online Reviews' page you need to follow the instructions to enter the URL of a page which has the data you require. We have some more information about retrieving online reviews which includes an informative video.
Analysing your data
If you have your own dataset that you would like to analyse or would like to analyse our sample file, you will need to create a project first. Add a new project by clicking on the ‘New’ button on the Explorer homepage or the ‘Create project’ button under the 'Projects' menu at the top of the page. Please note that if you analyse a project from our sample file, you will not be charged any credits for analysis.
If you would like to create a project using a dataset online that you have retrieved, you can create a project by clicking the ID of the reviews dataset and following the instructions on the screen.
Clicking either of these buttons will redirect you to a new page where you can begin the process of creating your project.
The Gavagai Explorer will guide you through the process of creating your project. At any point you can click on the 'EXIT' button to return to the homepage.
Header Row verification
For your analysis to be accurate, you need to ensure that your dataset has a "header row", ie, a row of headers for each of the columns in your dataset. The Gavagai Explorer will display the headers that is has detected and if they are accurate, you can move forward to the next step.
Naming your Project
At this stage, you can provide a name to your project. If you do not provide a name to your project, the name of the file will be used. Note that you can edit the project name later in the project's Settings page.
Select the Column to analyse
Once your dataset has been uploaded and you have named your project, you can select the column you would like to analyse. Gavagai Explorer will suggest the largest column as the column to analyse, but you can change that at this stage. Additionally, Gavagai Explorer will automatically detect the language of the column for analysis. If you feel that this language detection is incorrect, you can change the language that you use for analysis. Please note that the column which is analysed and the language used for analysis cannot be changed later. Click the 'EXPLORE' button when it is enabled, to proceed.
Exploration
Gavagai Explorer will then perform analysis of your data file and once the analysis is complete, you will be presented to view the results in a dashboard or perform modelling of the topics and groups to ensure more accurate results. You can also see your data in Explorer dataset
You can view your project in your home screen, along with your other projects. Next to the name of each project you can see it’s status, amount of texts and columns, owner, the date it was created and its current version. It will help you navigate between all your projects. There are also 3 dots to the right of every project, where you can perform actions like: Sharing the project, viewing its settings, or deleting the project.
You can also arrange your projects in different folders for easier navigation. You can add a new folder by clicking on the ‘New Folder’ button inside the 'New' menu.
You can then rename folders, drag-and-drop projects in and out, and delete them if they are no longer needed. Please remember that you can delete folders with projects in them, but the projects will be deleted as well and will not be recoverable.