Changes between Version 14 and Version 15 of BluePrint/Messaging/ExtendingParsing
- Timestamp:
- 07/11/13 13:07:42 (12 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
BluePrint/Messaging/ExtendingParsing
v14 v15 14 14 * Things that we want to extract and are essential requirements for the framework are discussed below. 15 15 16 == Data Model Changes == 17 18 * Make msg_message a Super Entity 19 * Each Channel would have an instance of this super entity which acts as the InBox and/or OutBox as-appropriate for that instance type 20 * The 'Master Message Log' then becomes the view of the super-entity (rather than having to copy messages here) 21 * Move non-core fields to component tables so that the core tables are uncluttered & fast 22 23 16 24 == Input Source Improvements == 17 25 … … 26 34 27 35 == Parser Improvements == 36 37 === Topic Detection === 38 39 * KeyGraph is used to detect topics across tweets/other feeds to filter relevant and actionable information from the rest of them. This is done after doing a loose filtering of information based on keywords and location. 40 * See http://keygraph.codeplex.com/ . 28 41 29 42 === Actionability === … … 43 56 * Implementing a UI which prioritises message parsing for starred senders is a useful requirement. 44 57 * The user should be able to *star* senders and *mark* keywords through the UI. 58 * Possible inspirations: 59 * dataLists with Filters (like TweetDeck): 60 * http://tldrmpdemo.aidiq.com/eden/default/index/updates 61 * S3Profile: 62 http://tldrmpdemo.aidiq.com/eden/org/organisation/1/profile 63 * S3Summary: 64 https://sahana.mybalsamiq.com/projects/sahanacommunityresiliencemappingprojectfinal/naked/Risk+Summary?key=ff49e93ddf8139e5eb61065660c796caa6f95845 65 http://i.imgur.com/jjaDmQ1.png 66 http://twitris.knoesis.org/indiarain2013/ 67 * Tweak the Tweet (EPIC) 68 * http://idisaster.wordpress.com/2013/05/28/more-research-on-boston-marathon-official-twitter-activity-smem/ 69 * https://github.com/ushahidi/SwiftRiver 70 * http://twitcident.com 71 * http://wis.ewi.tudelft.nl/twitcident/ 72 * Different Users: 73 * Power user looking for info themselves 74 * Dedicated miner (volunteer/intern/junior) mining stuff for decision makers 75 * Vision: be able to move between Filtered view & Firehose in gradual increments. Be able to train the automated assistants to make the filtered view more useful 76 (subscribe/unsubscribe) 77 * Features 78 * See all Messages in a datatable/list across media types (FB/Twitter/RSS/YouTube/Flickr) 79 * Filter them 80 * Add Sender to Whitelist/Blacklist 81 * Add Keyword to back-end filters 82 * View Images/Video 83 * Find Situation Reports 84 * ReliefWeb, etc 85 * Grouping/Linking results both to enhance validity & also provide a single point of entry 86 * Route to other Sahana Modules 87 * Drag and Drop between Raw source & Target Module 88 * Mark for Action 89 * create Tasks 90 * create Incident Reports 91 * create Assessments 92 * create Situation Reports 93 * Forward via Outbound Channels (Public e.g. Twitter & Private e.g. Email/SMS) 94 * Semantic Search? 95 * RDF Channel? 96 45 97 46 98 == Use Cases == … … 51 103 mis-typed / users moved / etc. 52 104 53 === Situational Awareness Dashboard ===54 55 * Trying to present a single,simplified view for many disparate information sources including RSS feeds.56 * Decision Makers have many different feeds of information flowing to them from both internal & external sources.Especially external sources may come in the form of RSS feeds . So, having the system somehow prioritise & classify things before they reach them (instead of manual screening) will make their task much easier.