2021-01-20: 366 dots in 2020 - top news stories of 2020

By -

Fig. 1 (Click image to expand) 366 dots in 2020 - Top news stories for 366 days in 2020. Each dot represents the average degree of the Giant Connected Component (GCC) with the largest average degree across all the 144 story graphs for a given day. The x-axis represents time, the y-axis represents the average degree of the GCC. The annotations (and legend) represented by colored dots were assigned semi-automatically.

I join the chorus to say 2020 was a year like no other, and shaped by three historic events: the Coronavirus pandemic, the protests surrounding the Black Lives Matter movement, and the US Presidential elections.

According to StoryGraph, in 2018, the top news story was the Kavanaugh hearings. In 2019, it was the Mueller Report. Similar to 2018 and 2019, we analyzed all news stories collected by StoryGraph at 10-minute intervals every day in 2020, to identify the top news stories of 2020. Recall how we identify top news stories, explained briefly in 365 dots in 2019 and in detail by the tech report, and repeated here for convenience.

Summary of StoryGraph's "top story" selection criteria

A "top story" according to StoryGraph is a story with a connected component with a high average degree. Connected components are generated by linking highly similar nodes (news articles) represented by the entities (e.g., people, locations) in the news articles.  The news articles were extracted from the RSS feeds of 17 US news sources across the partisanship spectrum (left, center, and right). The code is available and the algorithm is described elsewhere. But essentially, the more various news organizations use the same entities (e.g., people, locations, organizations) in their reporting, the more important the story. A graph is constructed with the nodes represented by entities extracted from news articles, and an edge between a pair of news articles nodes represents a high degree of similarity between the news articles. A connected component represents a news story (e.g., Mueller Report), and the average degree (y-axis in Fig. 1) of the graph's connected components (GCC average degree) is the attention score.

The top news stories of 2020

The top news stories of 2020 according to StoryGraph centered around the 2020 US Presidential election (e.g., Las Vegas Democratic Presidential Debate, Biden's Super Tuesday Wins), the US Supreme Court (e.g., Trump nominates Justice Barrett to the Supreme Court, Death of Justice Ruth Bader Ginsburg), the assassination of Iranian Commander Soleimani by a US airstrike, etc.

 
Even though Coronavirus was a constant in the news in 2020 --- as it permeated almost all activities from the elections to protests --- no Coronavirus story made it to the top five. In fact, the infection of President Trump with Coronavirus ranked 7th place. This might seem surprising at first; if Coronavirus was constantly in the news in 2020, why did it not rank higher than 7th place? A possible reason for this disparity is as follows. StoryGraph's top news story algorithm is sensitive to the magnitude (represented by GCC average degree) of attention news stories receive and and not the duration of the attention. This means the algorithm credits stories that garnered more attention without accounting for how long (hours vs. days vs. months) the attention lasts. The reason we opted for the magnitude of attention criteria over the duration of attention criteria is because it is a simpler approach. Accounting for duration is complicated by fact that it requires clustering stories across contiguous and non-contiguous temporal time spans among other factors. 
 
Tracking the duration of Coronavirus (orange dots) coverage by focusing on three event cycles. First, early reports from Vox about the virus spreading in China and beyond. Second, steady coverage from March to mid-May 2020, briefly interrupted by protests surrounding the death of George Floyd. Third, the return of steady coverage in October 2020 following the infection of President Trump and other prominent Republicans.

However, we approximated the duration of coverage of Coronavirus with a simple (and flawed) approach of counting the number of days with news articles that mentioned "Coronavirus" and it's synonyms. The same method was applied to approximate the duration of the coverage of "Protests." From Fig 1 and Fig. 2 (variant of Fig. 1), "Coronavirus" was top in the news for approximately 177 days and "Protest" - 39 days.
 
Top 10 Sumgrams in 2018, 2019, and 2020
The table below takes a different approach (from Fig. 1) toward summarizing the top news stories of 2018 -- 2020 by extracting the top 10 sumgrams (conjoined ngrams) for each year. The top sumgrams for each year were extracted as follows. First, we created a document (sumgram input) by extracting all (7,000+) titles from the news articles (nodes) of all top news story (from Fig. 1)  connected components. Second, we extracted 10 sumgrams (with base_ngram = 2) and removed stop words that consisted of the names of news organizations and synonyms (e.g., "president trump") of already represented ngrams (e.g., "donald trump").

Rank 2018 Sumgrams
 2019 Sumgrams
 2020 Sumgrams
1 white house
 white house joe biden
2 brett kavanaugh mueller report white house
3 donald trump impeachment inquiry supreme court
4 supreme court donald trump covid 19
5 michael cohen trump impeachment impeachment trial
6 north korea democratic debate donald trump
7 christine blasey ford joe biden amy coney barrett
8 john mccain el paso bernie sanders
9 stormy daniels trump indicted kamala harris
10 george bush nancy pelosi new hampshire

The colored text in the table represents ngrams present in 2020 and the previous two years. The table reveals the elevation of joe biden in the national conversation from 2019 (7th place) to 2020 (1st place). In contrast, the frequency of mentions  of donald trump steadily declined from 2018 (3rd place) to 2019 (4th place) to 2020 (6th place). The supreme court was prominent in the conversation in 2018 (due to the Kavanaugh hearings), it disappeared from the rankings in 2019, but reappeared in third place in 2020 due to the death of Justice Ginsberg, and the nomination of Judge Amy Coney Barrett to fill her seat.

The top 10 sumgrams of 2020 also featured an unwelcome guest, covid 19 (4th place), one we most certainly hope, takes the exit in 2021.

-- Alexander C. Nwala (@acnwala)

Comments

Post a Comment