IN-SPIRE Video Tutorials
All videos are in h.264 format and will either play natively in your browser or through the Adobe Flash Player. They can also be downloaded in QuickTime or Windows Media Video format.
IN-SPIRE Version 5
Getting Started with IN-SPIRE
An overview of the application and most-used tools (version 5.6).
- Duration: 4 minutes
- File Size: 50MB (.mp4 | .wmv)
- Topics: Project Manager, Dataset Editor, Galaxy, Document Viewer, Groups, Themes, Search, Facets
IN-SPIRE 5: Exploring Data Relationships with Facets and Themes
IN-SPIRE 5 has many advanced algorithms to process text. One of the best ways to use them is with the new Facet Tool. Learn how to slice your information into related chunks using the new IN-SPIRE Themes and how to create facets of your own.
IN-SPIRE 5: Using the Time Tool to Explore Temporal Patterns
IN-SPIRE 5 introduces a new tool to explore the date fields in your data. It graphically shows your documents, groups, and terms as they change over time. This powerful tool is a great way to find emerging and surprising topics
- Duration: 7 minutes
- File Size: 23MB (.mov | .wmv)
- Topics: Time, Term Details, Exploring Data, Groups
IN-SPIRE 5.8: Adding Domain Knowledge with Lexical Analysis
IN-SPIRE 5.8 adds a simple utility to turn lists of terms related to your domain into lexicons that you can measure your documents against. This video shows the benefits of lexical analysis and how to create your own lexicon lists.
- Duration: 8 minutes
- File Size: 50MB (.mp4 | .wmv)
- Topics: Lexical Analysis, Utilities, Exploring Data, Groups
IN-SPIRE 5.8: Phrase Boosting with the Advanced Settings Utility
IN-SPIRE 5.8 adds a simple utility to set some parameters that effect how IN-SPIRE extracts keywords and clusters documents. One of the options allows you to have IN-SPIRE favor multi-word phrases over single-word terms. Phrase boosting is useful in datasets with technical or specific terms in the documents.
Version 4.6
Note: The version shown in these videos is 4.6, but the techniques still apply to version 5.x.
- Creating a Visualization from a Google Search
- Creating a Visualization from Text Documents
- Using RSS Feeds to Create Real-Time Visualizations
- Creating a Visualization from Microsoft Outlook
- Creating a Visualization from PubMed
- Exporting and Importing Datasets
- Exploring Data in the Galaxy
- Creating a Subset
- Organizing and Comparing Documents using Groups
- Using the Correlation Tool to Compare Groups
- Viewpoints: Influencing the Visualization with Domain Knowledge
- Duration: 9 minutes
- File Size: 26MB (.mov | .wmv)
- Topics: News Feeder Service, Dataset Editor, Galaxy
- Duration: 4 minutes
- File Size: 18MB (.mov | .wmv)
- Topics: Dataset Editor; Adding source files; Using a template
- Optional Download: PubMed Template (zip file)
- Duration: 6 minutes
- File Size: 13MB (.mov | .wmv)
- Topics: Dataset Editor, Galaxy, ThemeView, QVP (Quick View Panel), Outliers, Document Viewer
- Duration: 4 minutes (.mov | .wmv)
- File Size: 18MB
- Topics: Galaxy, Time Slicer, Data Management, Dataset Editor, QVP (Quick View Panel)
- Duration: 10 minutes
- File Size: 27MB (.mov | .wmv)
- Topics: Groups Tool, Query Tool, Time Slicer, Document Viewer, QVP (Quick View Panel)
- Duration: 9 minutes
- File Size: 24MB (.mov | .wmv)
- Topics: Correlation Tool, Groups Tool, Time Slicer, Document Viewer
Creating a Visualization from a Google Search
IN-SPIRE can harvest web pages from a Google search using the main site, Google News, or Google Scholar.
Creating a Visualization from Text Documents
IN-SPIRE can harvest text files stored locally on your system, and can optionally utilize standardized sections in the text file to automatically extract categorical information.
Using RSS Feeds to Create Real-Time Visualizations
RSS feeds are a popular way websites can keep you informed of content updates. Many news organizations and blogs publish updates via these documents. This video shows how you can make a live visualization from Yahoo News feeds.
Creating a Visualization from Microsoft Outlook
IN-SPIRE can harvest text from Microsoft Outlook e-mail messages via a simple drag-and-drop mechanism. This is great for mailing lists or systems that send search results via e-mail.
Creating a Visualization from PubMed
PubMed is a free search engine for accessing the MEDLINE database of citations, abstracts and some full text articles on life sciences and biomedical topics. PubMed can download search results in a structured text format (MEDLINE), and IN-SPIRE can utilize the settings in template files to automatically extract categorical information.
Exporting and Importing Datasets
If you need to give a dataset (including all source documents, groups, and query history) to someone else, or to use on a different installation of IN-SPIRE, learn how to export using the Dataset Editor. This video also shows how to import an exported dataset.
Exploring Data in the Galaxy
When you first open a dataset, the primary method of exploring the information is the Galaxy visualization. Learn tips and tricks to help find out what's in a dataset.
Creating a Subset
IN-SPIRE can open datasets with multiple thousands of documents, but that's not always the most efficient way to do an analysis. Many times you want to look at just a chunk of the data. This video shows two ways to chop your datasets into smaller pieces (called subsets) for more detailed analysis.
Organizing and Comparing Documents using Groups
The Groups Tool in IN-SPIRE allows you to make collections of documents based on a selection, keyword search or automatically defined from categorical data. This video covers how to create and manage groups, and some powerful ways to use them.
Using the Correlation Tool to Compare Groups
The Correlation Tool in IN-SPIRE lets you view your groups in a grid so you can easily find and select interesting intersections. This video shows how to start using the tool and some tips and tricks to help maximize your analysis time.
Viewpoints: Influencing the Visualization with Domain Knowledge
IN-SPIRE is very good at clustering similar documents together, but it does not know the domain of the data. Using the Viewpoints tool in IN-SPIRE, you can give lists of terms "extra credit" in the visualization, creating clusters more in line with your domain.