Skip to main content
Fig. 3 | Journal of Big Data

Fig. 3

From: Large scale analysis of violent death count in daily newspapers to quantify bias and censorship

Fig. 3

Time profile of datasets retrieved from the newspaper archives. From left to right, top to bottom: NYT, CDS, REP, STA. For each newspaper is shown: (i) T(t): Total number of articles present in the archive (black), (ii) K(t): number of articles returned upon query of the word ‘killed’ or ‘morti’ (red), (iii) \(K_{f}\): Number of foreign events (cyan), (iv) \(K_{d}\): Number of domestic events (green), (v) \(R'= K_{d}/(K_{f})\): Ratio domestic/foreign events \(\times 10^2\) (purple), (vi) \(R=K/T\): (killed/total or morti/total) \(\times 10^2\) (blue). Red bands indicate the time of US Civil War (NYT), WW1 and WW2. It is possible to see the increase of T and K in NYT during Civil War and the corresponding decreases in STA and CDS during the two World Wars. For the same papers it is also evident a gradual decrease of K and \(K'\) during Fascist period (1923–1943) due to censorship (see text)

Back to article page