Anatella Plug-ins

Anatella offers two types of plugins:

  • Javascript-based plugins. Most of them are free. See what’s the community has to offer!
  • C++ based plugins.

1. Javascript-based plugins

1.1. Above-The-Line media-mix optimization

This plugin allows the Analysis of Time Series in combination with a predictive analytic analysis performed with TIMi.

This plugin offers a set of operators that allow you to find which Above-The-Line investment gives you the highest ROI. You will obtain Graphs&Charts that show, for each past weeks/months, the contribution of your different investments on your sales.

Investments can be of any nature:

  • Investments in Television, Radio, Outdoor display, Magazines, etc.
  • Investments at the point of sale (flyers,stands)
  • Investments in internet (SEO,banners,etc.)

The plugin can also exploit any other external data (other than your investments): for example:

  • The Investments from the competitors (if you have them)
  • Any external factor: temperature, humidity, country index, school holyday, etc.

2. C++ based Plugins

2.1. OLAP reporting

Anatella provides some OLAP reporting functionalities through the usage of a “Microsoft Office Data Injection operator”: This operator (i.e. “box”) allows you to automatically update any chart or graphics contained in any Microsoft Office document using some data extracted from any database(s) or flat file. For example, you can obtain, in a few mouse clicks, each day, automatically, an updated copy of your preferred PowerPoint presentation. Anatella support ALL types of MSOffice graphs: pie chart, 3D surface chart, bar chart, doughnut chart, bubble chart, etc…

2.2. Social Network Analysis (SNA)

This plugin offers a set of operators mainly useful for telecommunication companies (and also in a lesser extend for banks), to analyse churn (in combination with a predictive analytic tool). The objective of this operator is to extract out of the “phone- communication-network” valuable social-metrics.

Typically, the “phone-communication-network” is defined in this way:

  • Each individual (each subscriber) is a node.
  • An “arc”? of the network between the two individuals A and B represents the relation “A called B”.

The social-metrics that could be extracted from the network are: The best connected individuals, The individual who plays the most important role in any group (i.e. the social leaders), The groups of friends (the communities), The proximity to a churner, The number of churners in the “neighborhood” of an individual.

Those metrics can improve substantially the accuracy of predictive models for churn detection for telecommunication companies (and also for banks).

2.3. Binary CDR extractions

Anatella is able to read natively the Binary CDR (Call Data Record) files generated by common Telecommunication hardware such as the Ericsson 100313-1505, various Ericsson GGSN, various Ericsson SGSN and various Siemens D900. Anatella offers one of the fastest ASN.1 decoder available on the market. The Anatella ASN.1 decoder is 100% compliant with the Abstract Syntax Notation One (ASN.1) Formal Description, that is specified in the International Telecommunication Union (ITU-T) Recommendations X.680-X.683 and X.690.

The ability to work natively with raw CDR files is great for datamining purposes because you have direct access to all information available in the original CDR files. Usually you only get, for your datamining analysis, a sub-set of the CDR file information. The rest of the information is lost during the standard data management procedures tha are typically implemented by a telecommunication operator. For example, Anatella can extract more than 970 features out of the raw CDR files create with an Ericsson 100313-1505. This is a tremendous opportunity for a dataminer!

2.4. Automated text-spelling correction

Click here for more information about this plugin.

2.5. Text-mining

These operators apply the classical “bag-of-word” technique to produce, starting from raw, unstructured text-data, many new columns and new variables directly exploitable inside TIMi or Stardust, for predictive analysis.

You can now exploit the information contained inside any unstructured text-data (for example: text comments on your website, news articles, …) to make any predictive model (for example for cross-selling, upselling and churn)!