I read the 1st third (it's really long) and while the data analysis is interesting, the conclusions say a lot more about the biases of the author(s) than those of the BBC.
Fundamentally you can't use sympathy as a measure of bias without first establishing a baseline for how sympathetic the views and/or groups of people are. The report mentions that Palestinians might be more sympathetic because they're the ones being blown up, but then discards this by pointing out that the BBC is supposed to "ensure broadly comparable treatment of the Palestinian and the Israeli viewpoints" without acknowledging that maybe they do and one viewpoint is more sympathetic than the other. The least sympathetic group according to the report is Hamas, so according to it's logic they're the group the BBC is most biased against. Not a reasonable conclusion.
There's plenty of other indicators that this report started with a conclusion then tried to gather data to support it, but I've already spent more time on this comment than the report deserves.
YZF|10 months ago
I think it's an interesting question of how we measure bias.
For me, as an Israeli (who hasn't lived there for decades), who has some first hand knowledge of the situation, much of the reporting appears to be extremely biased. I know there are claims from the other side the bias goes in the other direction. What's the ground truth? I think using AI to crunch the large amount of data is a decent first order approximation.
Ofcourse bias depends on ideology. For some people if a Palestinian guns down an Israeli in a Tel-Aviv bar simply reporting this fact is biased towards Israel. And I mean, from their position that is understandable. And indeed we can see some media outlets that would not report these events at all, which I would consider an anti-Israeli bias.
quails8mydog|10 months ago
I'd also agree that using AI for sentiment analysis could be a good approach, I'm not an expert in the area, but I believe this is one of the things AI is best at. But it needs an extra step to translate that into bias. Establishing a sympathy baseline is my initial idea, but I haven't tested it and maybe there's something better.
Whether something is biased is less about how any given individual(s) feel about what's been said and more about if the different viewpoints are presented honestly. Though it can get really difficult to identify except in the most extreme cases. As you say, it's not just what's said where the bias occurs, but also in the choice of what not to say.