Analysis and Visualization of Big Data
Chicago Crimes Map
Newspaper and journal editors have to read and classify many crime texts provided by local police stations every day. Several and various parameters categorize the crime texts and then pushed into databases for further processing. This manual process is very time-consuming. However, it is an essential area of editorial work, as this text processing reveals statistical analyses on frequency distributions of several crimes as well as occurrences of specific crimes in a given period or city areas. Editors can use this information to find crime hotspots for exciting news stories.
We use specified heuristics as well as artificial intelligence in our text classification system, to automatically classify crime texts according to the customers’ needs. In our current version of this POC, we extract text information such as the crime category (Theft, traffic accident, …), the date when the crime occurred, as well as the place/street of the crime. This process is done entirely autonomous and in real time. New crime texts are sent to our cloud where they are processed immediately to be stored in the database.
Our Crime Map recognizes and then lists out crimes in real time onto a city map. Thus, users can instantly find certain crime hotspots in a particular area of the city, categorized by the origin of the crime. This solution also offers the direct search of a felony by using keywords. The Crime Map Solution does not only help newspapers and journalists who are always on the hunt for a new and interesting story, but it can also be helpful for private end users. Users, e.g., which are looking for a new apartment or home can use the database to search areas for occurred crimes. Not only the crime itself can be looked up but also the frequency crimes occur in a particular area.
To improve this POC, we are currently working on an additional feature, which can extract the categorized information of the database to add an additional system based on Artificial Intelligence (AI). That AI-driven systems then detects anomalies in pre-forecasted crime rates. As soon as the system identifies potentially interesting anomalies, a trigger is sent to the editors/newspapers with the suggestion of a breaking news story.
To sum up, our crime map with integrated AI based text classification provides full insight for a specified target audience to extract respectively relevant information. Furthermore, our system helps editorial processes regarding time management and story-finding. Thus, editors will have more time to focus on their critical tasks, such as writing interesting news articles.