Huginn's News believes that users should be able to actively analyze news. For this, Huginn's News provides online graphical analysis tools. Huginn's News believes that if users interact with their news, then users may be more critical and come to more interesting conclusions. For example, the source of an article is not immediately clear. It is possible to display the source of an article, but users must toggle that feature. Articles are grouped together based on content and not source.
Huginn's News has a minimal amount of human interaction during the preparation stage. Currently the only human interaction is choosing which sources collect, and the choice of algorithms. All of which are clear upfront. Since a computer algorithm has not political affiliation it is assumed to be less biased. Huginn's News writes no articles itself.
It is well documented that news is predominantly negative. Several factors can contribute to this. Including the human tendency to react more strongly to negative news, as well as journalists tendency to negatively question policitcal news. Huginn's News believes that new organizations select subject matter that fit an agenda, and simply will not report stories not fitting their agenda. This presents a bias in news reporting, and is not specific to a single news outlet.
It is assumed that many people tend to consume news from a few websites. With people's attraction to negative news and a tendency to ingest a single source of news, this will further polarize peoples opinions, misinform them, and can be depressing.
One problem is that there is no central site combining tools to analyze news with stories aggregated from different sources. Huginn's News believes that if people interact with their news, then they may see several sides of an issue.
Huginn's News' solution to these problems is to bring data from different sources to a single location, and present tools allowing users to view news differently. The source of each article is not immediately obvious to the users, but can be found with a few clicks. News articles are aggregated by content, updated hourly, and analyzed hourly.
Huginn's News presents a solution that combines Machine Learning and automated data acquisition. Huginn's News performs analysis using Natural Language Processing (NLP) to determine subject matter. Articles are matched using Cosine Similarties, and topics are extracted using Term-Frequency-Inverse-Document-Frequency. Note that this is all done without human interaction, introducing a minimal human bias in the analysis.
Users must select different features when reading the news, and interact with the website. They can select positive news or different sources, but not both at the same time. Huginn's News hopes that by having a single platform for users that includes multiple sources and analysis, people will be better informed.
If you would like to include a source, then let Huginn's News know.
If you would like to help, then please contact Huginn's News.
I am an Engineer and Analyst. You can find more information about me at my My LinkedIn Profile.