Getting started with Gavagai Explorer
Quick video version on Youtube.
Gavagai Explorer is a tool to analyze related texts to find common topics, their associated terms, their sentiment scores and their importance (in terms of number of respondent mentions). Although the primary use case for Explorer is open-ended question analysis of surveys it can be used for the analysis of any set of related texts such as hotel reviews.
Gavagai Explorer aims to give the analyst a comprehensive and quantified view of the data; it identifies the main topics or topics and measures their importance as a relative strength to the total number of texts. This means that the strength of a topic is comparable across data sets (and this type of tracking over time will be provided in a future version of Explorer).
This tutorial introduces the main features of Gavagai Explorer:
- Importing your initial data and filtering it to your needs
- Viewing the data and moving around
- Editing Topics: Grouping and Merging
- Ignoring topics
- Sentiment analysis
- Exporting data to PDF and Excel
- Models: Saving and Applying your Models to other Projects
Create a free account here: https://explorer.gavagai.se/#register
Importing your initial data and filtering it to your needs
When you press this button you will be redirected to a new page where you can give a name to your project and load your data. Write the name of your project in the text field. Note that you can edit the project name later in the project Edit page. Load the data to be analyzed by clicking the Choose file button and select a csv file (comma separated file with data in columns) or an Excel file (either .xls or .xlsx).
When you choose your target file, you will be redirected to Explorer homepage (My Projects page) where you can see the list of your projects including the one you just created. Wait a few seconds until Explorer completes pre-processing your data and the Status of your project turns to Ready. You can then click the Explore button to start analyzing your data.
You first need to specify the main text column that you want to explore. Note that if your file consists of more than one text column that you want to explore, you need to create a separate project for each of these columns.
After loading the data you can also add filters to select a subset of all rows for you current analysis session. This requires that you have columns in your file that contains meta-data. Select such a column, in the example below we select a column containing the rating value for hotel reviews (where the review text itself is in the main column). Gavagai Explorer then presents the possible values from this meta-data column to let you select which values you want to include. You can add as many filters as you need.
Viewing Data and Moving Around
To quickly jump to the top of the page, click the go-to-top arrow button.
To see the complete text instead of just the parts that are relevant for the current topic, click the Show original text link in any text.
Editing Topics: Grouping and Merging Topics
Topics are the main topics found in the data when analyzing with Gavagai Explorer. It is sometimes useful to structure the topics according to your own idea of how things relate to one another. With Explorer this is easy by using the features.
You can edit the topic name without changing the actual topic terms or their synonyms:
When you hover mouse over a topic in the working panel, an upwards arrow appears inside the topic box which is called Select button. You select a topic by clicking on this button. When the topic is selected, the topic blue box turns to orange.
In Explorer, both merging and grouping should start with selecting a topic. In the following it is explained how we can merge and group topics by Explorer.
For instance, for the hotel reviews, you might like to put all the words about food (e.g. breakfast, coffee, wine, etc.) in one group called Food. Explorer shows you the number of the respondents that have written about food (frequency of the whole group). It also shows you the aggregate sentiments of this group, so it will let you know if the respondents are positive or negative about food on the whole.
When you click on this button, a down arrow button will appear in the left side of each topic or group.
For instance, for two topics income and compensation including the terms income and compensation successively, you might merge them to get a single topic including both the terms income and compensation. Merging topics increases the strength of the resulting topic in terms of the number of texts that are included in the topic and therefore it results in preciser sentiments and more informative associations.
Similar to grouping topics, merging topics starts with selecting a single topic. As before, you select the topic by clicking on the Select button in the topic box. When the topic is selected, an upwards arrow button will appear at the left side of all other topics.
You can remove irrelevant topics from your analysis by ignoring them. If you ignore a topic that topic is no longer taken into account when clustering responses, but the texts themselves are still available for inclusion in other topics. To see the effects of the ignore you need to re-explore.
For every topic the system calculates the most relevant associated terms. You can look at these by clicking Show associations.
When you click an association the corresponding texts of the topic are filtered by that association resulting in a smaller set of documents on the right hand side.
When you download the result of your analysis as Full Excel or Full CSV, you will have a Excel or CSV file containing your original data appended by the analysis result from Explorer. Explorer performs a row based analysis of your data in respect to your pinned topics, target concepts and sentiments, and for each text it adds the result of the analysis to the corresponding row. In the following we explain the appended columns.
And the following shows exporting to Excel:
Models: Saving and Applying your Models to other Projects
You can save a project in Explorer as a model and then apply the model as a template for other projects. Using models is a suitable approach when you analyze similar types of data frequently and you have same concerns in your analysis; for example,e.g. you are looking for the same topics or topics. Here we explain models in Explorer and we guide you through creating and using them.
A model consists of your specific settings in a project; meaning grouped topics, merged topics, pinned groups and topics, and your ignored terms. When you apply a model on a project, the project current model will be replaced by the new one. Therefore, if you are unsure, you would better save the current model first.
Note that to apply a model to your project, the project’ file does not need to have the same specifications as the model’ source file (e.g. same columns or same number of texts, etc.).
You can create a model from a project by selecting Model under Save as drop-down menu in the working panel.
At the top of the project, you can click the drop down box to apply models. The full list of your models can be found under My Models page.
If you would like to get more information about all the features and functions, have a look at full Gavagai Explorer Documentation here.