|
The technology analyzes documents using a Natural Language Processing engine. This allows the document to be “read” not just analyzed. This analysis results in a document being scored by a few standard-rating systems.
- The Gunning-Fox Index
- Linsear Write
- Flesch Reading Ease
- Flesch Kincaid Grade Level
- SMOG Index
- Automated Readability Index
- Coleman-Liau Index
These rating systems allow the analysis to build a profile or the target audience and the writer. As a result these indexes estimate the average reading level of the target audience.
In addition to this analysis a custom positive impact rating scale is analyzed. Each sentence is analyzed as a positive negative or neutral statement. These statements are then applied to a target. (Usually the subject of the article but it can be a person, place, or any other proper noun). The result of this algorithm is a score based upon all major or significant people, companies or places in the article. This score tells us if the article is in favor of a specific topic or against a topic.
The final stage of analysis before indexing and mind mapping is Information Extraction. This compiles the most common and important information from an article such as key places and people. An example article and the extracted information should be accompanying this description.
Once the information is analyzed it is then indexed for searching and mind mapping. This process extracts and categories all of the important and key words from an article. It also then creates a full text index of the article.
After indexing a document has a mind map created from it. This allows a users using the mind mapping software to read an article and click on any mapped nouns or verbs and be presented with a dictionary definition, common aliases (or synonyms). It then will present a cross referenced list of linking material. As an example if you are reading an article and the name of a film director is in it. When you click that film directors name you are given a list of movies they have directed or appeared in, articles written or containing that authors name, the current analyzed public opinion of that author, the size of the indexed database about that person, and any bio graphical information indexed about said person.
Memes, Hive Minds and other strange sounding things affecting mindshare…
Algorithmic Ventures offers unprecedented insight into online contextual meaning enabling unparalleled analyses of the genesis and propagation of ideas, thoughts, beliefs, etc. – also known as “memes” (rhymes with “gems”) - across the digital landscape. A meme can spread from mind-to-mind similar to the transmission of genes within a family over many different generations (however memetic proliferation can literally happen at the speed of light via the internet). Logically, a meme becomes most virulent when propagated within a group – the larger the group becomes the more prevalent the meme turns out to be. A group, in this case a collection of internet users, is generally called a meme-complex or “Hive Mind”. For the unacquainted, a Hive or Group mind is one in which any large group or population aligns around shared thoughts and/or beliefs. These thoughts or beliefs can quickly grow or shrink and the Hive itself can swell in membership or quickly perish. It is important to note that this dynamic, multidimensional “shared mind” is one in which the group characteristics are valued over individual identity. Given the internet’s capacity for easily disseminating ideas to the masses while allowing for a high degree of anonymity, the ability to monitor and analyze the internet “Hive Mind” en masse is of great value to anyone interested in understanding the raw power of shared opinions.
A depth of knowledge around shared online sentiment is a glorious thing indeed!
Imagine the significance of situational awareness afforded from conducting crawls of online conduits to monitor the "hive mind" around any meme, such as, "what does the online community think of Al Gore’s environmental stance?" From this a multidimensional score around collective sentiment can be constructed. If this analysis is run over a time series (hourly, daily, weekly, monthly, annually, etc.), we are able to actively chart a "memetic lifespan" to show how pervasive a meme is and how shared sentiments around ideas, thoughts and beliefs both live and die – this can produce important insight around predicative (future), current and historical (comparative) implications for public releases, political campaigns, marketing efforts, corporate identities, crisis management, corporate earnings releases, etc.
|
