Direkt zum Hauptinhalt

Machine Learning and Metadata Tagging (M2TS) Services

Tech Mahindra’s “Machine Learning and Metadata Tagging (M2TS) Services” provides Content Recommendation, Automatic Tagging of Content, and Content Organization and increase the findability of Related Content.

Home >> Partner >> Alfresco Partner Solutions Showcase >> Machine Learning and Metadata Tagging (M2TS) Services

Tech Mahindra’s “Machine Learning and Metadata Tagging (M2TS) Services” provides Content Recommendation, Automatic Tagging of Content, and Content Organization and increase the findability of Related Content. Using the Machine Learning and Big Data, M2TS builds topics and corresponding tags for each topic based on a common grouping and prepare Meta Tags. Once a document is imported in Alfresco folder which has integration with M2TS, the “Machine Learning tags” are auto populated to the document’s tag and establish a contextual relation.

Business challenges

The digital content is growing exponentially ~2,000 Exabytes added in 2016. For large Enterprise organization, finding the right content at the right time make a huge impact on critical decision making and business deals. However, if the documents are not organized and metadata are not tag/defined properly, then it became a nightmare for the DMS users to find the right documents.

  • Inconsistent document organization
  • Tedious and time consuming manual process to find a document and define tags for each document.
  • Insufficient metadata to find it easily
  • Human error
  • Difficult in organizing large volume of content

Solution Overview

  • To make search as contextual and make more relevant, solution was built in big data/Hadoop with Machine Learning technique for automated and reliable content tagging.
  • Enterprise search for indexing about 100 TB of structured and non-structured data.
  • Open and generic rest API provided for faster integration and automated testing
  • Auto classification done to tag the content according to security policies using big data with machine learning algorithm
  • Content clustering implemented inside big data with machine learning algorithm that finally goes inside HIVE Database for faster traversal
  • Service based integration integrations with Alfresco repository using the M2TS REST Web services.

How did the solution address the business problem?

  • Saves time , Increase Productivity
  • Provide relevant search results quickly
  • Automated tagging
  • Content Contextual recommendations
  • Smart  and Intelligent decision making with self-learning ability
  • Improve user experience
  • patterns recognition and organize the content

Customer Benefits

  • Improve productivity: Reduction in document search time
  • Improve knowledge management.
  • Effective and efficient process management.
  • Cost savings by eliminating manual efforts for classification.
  • Gain competitive advantage by adapting to the changing knowledge economy.

 

Weitere Informationen

Licensing Model:N/A
Alfresco Content Services:5.1, 5.0
Alfresco Process Services:5.1, 5.0
(System-)Anforderungen
Requires Additional Server Software Install We can use Cloud era pre-defined environment or Manual setup
Software specifications - For Cloud era:
- Virtual box CDH -5.0 (64 bit)
- Java -1.7
- Hadoop- 2.3.0 or later
- Hbase- 0.98
- Mahout-0.9
- Apache tika-1.5
- Eclipse
Need to install/use the below softwares/services
- Java -1.7 or Later (64 bit)
- Winscp-5.5.2
- Restful Web services
- Tomcat-6.0 or later
- VM Player -12
For Manual:
- Hadoop- 2.3.0 or later
- Hbase- 0.98
- Mahout-0.9/mallet 2.0.7
- Apache tika-1.5
- Luna Eclips
Hardware Specifications:
- OS (Linux (any flavor)/Ubuntu)-64 bit
- RAM- 8GB
- To get optimum performance, we recommend the following:
- 64 bit hardware and a 64 bit JRE
- Use a system with a clock speed above 2.5 GHz
- Minimum of 4GB RAM is required to install and run Alfresco
- Content in Alfresco is by default stored directly on disk, therefore to hold 1000 documents of 1MB in size will require 1000MB of disk space
- For 100 concurrent users or up to 1000 casual users: 1,5 GB JVM RAM
- 4x server CPU (or 2xDual-core)
Language(s):English
Version History:N/A
Support:Supported by Solution Provider

Kontaktieren Sie Alfresco

Wird geladen