Meaning Based Media Intelligence

Together with Autonomy, the leader in meaning-based computing and the world's fastest growing software company, VMS is developing next-generation media analysis and competitive intelligence solutions to provide PR and advertising professionals with actionable insights from “unstructured” data formats, such as social media, speech and music files on the web, radio, broadcast media and more.

Autonomy’s powerful suite of technologies extract meaning from these articles, online posts, videos and conversations and automatically performs detailed analysis to identify emerging trends and their implications for companies and products.

The foundation of meaning-based computing is a process called conceptual searching.  This unique capability allows users to search huge volumes of data using keywords and identify the ideas or “concepts” that are present in or related to the matching coverage.  For instance, searching on “dog” using conceptual search will return information related to “dog” such as "hound", "walks" and different breeds of dog, because the software understands the idea represented by the word.

Conceptual search is important because it locates critical, relevant data that might be missed by traditional keyword search and because conceptual search identifies related concepts and topics which can provide users with a better understanding of the key features of discussion across the media on products, services, issues, companies or competitors.

Following are just a few of the ways Autonomy and VMS help you derive new, deeper intelligence and meaning from your media coverage:

Heat Map 

The "Heat Map" presents the combination topics of coverage,  the amount of coverage and the sentiment or "heat" of that coverage.  It can analyze tremendous volumes of  coverage and present a snapshot image of topics of conversations and the related heat. Answering questions link,  "What are the major features of my media coverage and how are they being presented?"

Hot spots and colors show the convergence of coverage volume and positive/negative heat.  A large green circle indicates a signficant amount of positive coverage and the actual coverage that can be reviewed by simply clicking on that circle.  Conversely, a large red circle would indicate a topic with a significant amount of negative coverage.

Ideas Cloud 

The “Ideas Cloud” analyzes media coverage to reveal key topics and related items.  Once you do a search, a “cloud” is formed displaying all the key concepts taken from each “document” in the search results.  The larger and bolder "ideas" indicate the appearance of that concept in many documents in the search results. You can use the ideas cloud to refine or expand your search even further. Clicking on an idea refreshes the cloud with a more specific one that contains that new concept.   

Spectrograph 

The “Spectrograph” builds on the “Tone Map” concept.  The “Spectrograph”  displays coverage by topic over time with the related sentiment of the coverage.  This allows you to identify changes in the topics, volume and tone of coverage over time through the color and width of the lines on the map The brighter colors indicate "what's hot" while the width of the lines is an indication of the volume of information on that topic.  As a topics change over time, lines may split, converge or new branches can develop.  By scrolling over the spectrograph, titles for the topics are displayed and clicking on the line displays the coverage related to a particular topic .  

Cluster Map 

A cluster is a group of documents – clips, articles, Blog entries, etc – that are similar based on the “concepts” contained in each document. Clustering allows you to look at a lot of data and automatically identify themes or information clusters. Clusters that are similar are close together, and clusters that are less similar are placed further apart. You can track the coverage of your company that the getting the most attention as well as what does not – no matter how large the volume of coverage and across all media types.