Merge pull request #74 from CatalystCode/c-w-patch-1

Remove outdated Kafka entry

README.md

@@ -13,82 +13,7 @@ This project contains a Spark Streaming job that ingests data into the Fortis sy
 3. Narrow down the stream of events based on user-defined geo-areas, target keywords and blacklisted terms.
 4. Perform trend detection and aggregate the metrics that back Project Fortis.
 
-At the end of the ingestion pipeline, we publish the events to Kafka from where any downstream processors or aggregators
-can consume the data. The schema of the data in Kafka is as follows:
-
-```json
-{
-    "title": "FortisEvent",
-    "type": "object",
-    "properties": {
-        "language": {
-          "type": "string"
-        },
-        "locations": {
-          "description": "The ids of all places mentioned in the event",
-          "type": "array",
-          "items": {
-            "description": "A Who's-On-First id",
-            "type": "string"
-          }
-        },
-        "sentiments": {
-          "type": "array",
-          "items": {
-            "description": "Neutral sentiment is 0.6, 0 is most negative, 1 is most positive.",
-            "type": "number",
-            "minimum": 0,
-            "maximum": 1
-          }
-        },
-        "keywords": {
-          "type": "array",
-          "items": {
-            "type": "string"
-          }
-        },
-        "entities": {
-          "type": "array",
-          "items": {
-            "type": "string"
-          }
-        },
-        "summary": {
-          "type": "string"
-        },
-        "id": {
-          "type": "string"
-        },
-        "createdAtEpoch": {
-          "type": "number"
-        },
-        "body": {
-          "type": "string"
-        },
-        "title": {
-          "type": "string"
-        },
-        "publisher": {
-          "type": "string"
-        },
-        "sourceUrl": {
-          "type": "string"
-        },
-        "sharedLocations": {
-          "description": "The ids of all places explicitly tagged in the event",
-          "type": "array",
-          "items": {
-            "description": "A Who's-On-First id",
-            "type": "string"
-          }
-        }
-    },
-    "required": [
-      "id",
-      "createdAtEpoch"
-    ]
-}
-```
+At the end of the ingestion pipeline, we publish the events and various aggregations to Cassandra.
 
 
 ## Development setup ##