French Election ‘17: First Debate March 20th


Right Relevance (RR) provides curated information and intelligence on ~50 thousand topics. This includes:

  • Topic relationships including related topics & semantic information like synonyms.
  • Topical influencers (~2.5M) with score and rank.
  • Topical content and information in the form of articles, videos and conversations.

Additionally, Right Relevance provides an Insights offering that combines the above Topics and Influencers information with real time conversations to provide actionable intelligence with visualizations to enable decision making. The Insights service is applicable to emerging events like elections, conferences, product launches, breaking news developments, outbreaks like Ebola etc.

This report is a summary of graph analysis of engagements and conversations including retweets, mentions and replies of tweets from 8PM UTC (9PM Paris) on March 20th to 11AM UTC (12PM Paris) on March 21st, covering the chatter around the first debate for the French Presidential Election 2017. It builds upon learnings from the initial insights report posted on March 14th.

Data & Duration

The report leverages tweets sampled from 8PM UTC (9PM Paris) on March 20th to 11AM UTC (12PM Paris) on March 21st and along with Right Relevance topics, topical communities’ and articles data form the basis for the analysis.

The phrases used for gathering tweets are: “frenchelection”, “frenchelections”, “french election”, “france election”, “#france2017”, “#Presidentielle2017”, “#Frexit”, “Fillon”, “FrancoisFillon”, “Le Pen”, “LePen”, “MarineLePen”, “#Marine2017”, “#hamon2017”, “Hamon”, “BenoitHamon”, “macron”, “#macron2017”, “Melenchon”, “LeGrandDebat”

Most of the summary report is extracted from the analysis collateral in the form of:

  1. Tableau Online Dashboard
  2. Gephi Communities Graph Visual: Extracts are shown below.

For access to the complete graphs please send email to

The analysis methodology is outlined at

Communities Graph- Who Won the Debate?

Community detection graph algorithms like Walktrap and InfoMap are used to identify communities (as sub-graphs) in our engagements graph built using Neo4j & R. Graph visualizations are done via Gephi.

Wrt to analyzing the first debate, our focus is on the Retweets-only graph (Fig 1) since it is usually much better at bringing out differences.

The retweets-only graph (Fig 1) shows multi-polarity like our early analysis last week. Stark and clear communities can be seen for the 5 leading candidates in the French Presidential Election 2017. All 5 major candidates (Marine Le Pen/Yellow, Francois Fillon/Blue, Emmanuel Macron/Green, Benoit Hamon/Red and Jean Luc Melenchon/Orange) have their active engaged support bases that are distinct.

Figure 1: RTs-only Gephi Graph for French Election Debate 1

Apart from the 5 primary candidates, we can observe a separate subgraph in Pink that’s a community not related to any of the candidates.

So, Who Won the Debate:

At a high level, from the RTs-only graph, engagement numbers suggest Macron and Fillon led the debate and were neck to neck with Macron having a slight edge.

Marine Le Pen was next esp. including her English language engagements. Melenchon pipped Hamon with engagement numbers slightly lower than MLP. Hamon was last with half the engagement numbers compared to Melenchon.

But, the biggest winner of the night, with the most engagements, seems to have been parody.

Main Observations

There are seveal other points of note that can discerned from the graph, the metrics and Tableau which are outlined below.

Parody- Biggest Winner of the Night

Comedy seems to have been the biggest winner of the night with some of the most engaged tweets and highest ranked accounts for the debate being related to parody.

Some examples are below.



Figure 2: Top Tweets are mostly Comical

These accounts like @SeriousCharly@Scipionista@Sylvqin@le_gorafi along with many others form a highly engaged community (Fig 3). This community is parody heavy and has the highest engagements for the first debate. Much higher than any one candidate. It stands out as the largest community.

Figure 3: Parody heavy community is largest for Debate 1

@SeriousCharly and @le_gorafi also head two of the most prominent flocks that can be seen via Tableau. Both flocks are heavily infused with comedic tweets with broad engagements across the board including the presidential candidates’ communities.

Figure 4: Top Flocks for Debate 1 include two focused on Parody

Marine Le Pen Impact

Marine Le Pen (MLP), representing the far right, is attempting to repeat the success of Trump and Leave (Brexit) campaigns by trying to align with them along with espousing many similar goals. Interestingly, our analysis isn’t showing the same quality and quantity of engagements for her as seen in those campaigns.

This Bloomberg article notes the social media efforts by MLP so it’s possible that we’re not seeing supportive numbers due to multiple reasons.

  1. Potentially a large following for Marine Le Pen is older and rural and not as active on Twitter.
  2. Considering the far right nature of her platform, it’s more extreme that Trump and Brexit in some ways, leading to many supporters being less willing to express their views even online.

Also, another very noticeably issue is that only MLP seems to have a small but highly active community in English. We noted two extremely interesting accounts in our prior anaysis; @PrisonPlanet (Paul Joseph Watson, Infowars Editor-at-Large) and @vivelafra (ViveLaFrance); and decided to keep an eye on them. In this analysis, the English brigade for MLP forms it’s own distinct community (Fig 5), tightly connected to the MLP community.


Figure 5: “English” language community for MLP, Trump/Brexit mix

Many accounts which were very active in Trump’s election and Brexit, seem to have come together to pitch for MLP in the French election. As of now their effort, though noticeble, doesn’t seem to be breaching the broader French election conversation and forms it’s own echo chamber on the side. Their impact is likely higher the UK and USA but probably not in France.

Left/Center- Undecided Voters?

The undecided voters theme was significant in our early analysis and is noticeably, probably to a lower extent, as part of the debate analysis.

As seen in Fig 1, the  communities that comprise the “left” and left of center spectrum (Macron/Green, Hamon/Red & Melenchnon/Orange) show more engagement with each other with a large diffuse area with merging of colors around the center. These show a potentially large base of undecided voters esp. towards the center and left of center.

There is a big gap showing very light engagement with the “right” of center and far right. The MLP & Fillon groups seem more tightly knit. Their relative size and engagement number though are in line with center/left, esp. with Macron, showing that no-one has a clear edge

The above observations, along with the fact that parody saw the highest engagements, can reasonably reinforce the conclusion that uncommitted voters form a significant part of the electorate. This theme is similar to our US election and Brexit analysis and something to continue to keep an eye on as we go along.

Influencers-only Graph: Left/Center Heavy

The influencers-only RTs graph (Fig 6) shows a very different picture than the all users engagements graph.

Figure 6: RR Influencers-only RTs Gephi Graph


The first and foremost thing to notice is that, unlike the all users graph, the influencers graph is much heavier and far more prominent towards the left and center with the right being much smaller in size. This skew potentially points to an overall left leaning polity in general. This is fairly similar in nature, and even more pronounced, to the US election and Brexit . In both those campaigns, influencers were much more loaded on the Hillary Clinton and Remain sides, which are closer to the left/center idealogocially in the context of the French election.

One main exception visually to the above conclusion is the MLP-aligned english language focused community, which seems much larger in this context than the all users graph. This skew is due to the heavy english language focus of the Right Relevance platform for influencers. Also, several right leaning Trump/Brexit influential supporters have aligned themselves with MLP.








Write a Reply or Comment

Your email address will not be published. Required fields are marked *