500,000,000 - the number of posts made on X in ONE day.

1,500 - the number of posts that you are shown on the feed.

Talk about picking a needle from a haystack; that’s a 0.0003% filtering rate.

X, like all Social Media Intermediaries (SMIs), earns money from advertising. Advertisers, however, invest in X only when the company guarantees extended user engagement for their ads.

And that is the algorithm’s job: To keep you engaged as long as it can.

And for that, it has to show content ‘relevant’ to you.

The ‘Following’ tab is simpler: a chronological arrangement of posts from accounts you follow.

The ‘For you’ tab: The algorithm chooses posts that you’d probably like to see.

And that probability is the crux of the entire algorithm.

A Quick 5-min Tour: Demystifying the Complex in a Few Bites

Nerdy Nirvana: A Deep-Dive Into The Algorithm

Actionable Intelligence (Takeaways)

A Quick 5-min Tour: Demystifying the Complex in a Few Bites

The Process:

Now X has a fair understanding of Alex through his:

1. Likes, reports, quotes, followings, bookmarks, post clicks, time spent on each post, profile clicks, and more information like Alex’s geolocation, the trending topics in that location, etc. It also understands the content he does not like - collectively, this information is known as ‘Data’.
2. X assigns a different weight to each of these data. For instance, when Alex likes a post from one user (say Bella) and reposts a post from another user (say Charlie), X understands that Alex has more affinity towards Bella’s posts than Charlie’s. This happens on a large scale with Alex and every user and every post that he can be associated with. A graphical representation of the same is made. Such kind of a relationship is called ‘Feature Formation’.
3. X gathers 1,500 posts from the features so formed. In the next step, X tries assigning a rank to each one of them. The data comes into the picture again. Based on Alex’s interaction so far on X and based on the current popularity of each post in the 1,500 posts, X determines the probability of Alex - liking it, commenting on it, reposting it, etc. Here again, each probability is assigned a different weight. For instance, if Bella makes two posts and Alex is more likely (probability) to like one and comment on the other, the later post will be shown first. This process is known as ‘Ranking’.
4. Now that the posts are ranked, they undergo refinement. For instance, Alex will be bored if he sees 10 posts of Bella, one below the other, when he follows 100 other users. Alex should also not see content that is legally banned in his country. There should also be a good mix of posts from accounts that Alex follows and ones he doesn’t at present. This refinement happens in the Filtering, Heuristics, and Product Features stage.
5. The final step is to mix these posts with Ads (that’s obviously how X makes money) and follow-recomendations. This is the mixing stage.

The key areas where much of the work happens is in stage 2 & stage 3, i.e., Feature Formation and Ranking

For that, we’ll look into these 2 stages with more diligence considering two users viz., you, a post author who likes to maximize reach, and Alex, a random user on X.

Feature Formation:

Essentially, what happens here is:

Imagine you are a post’s author. Your intention here should be that both your profile and your posts have a strong relationship with Alex (in fact, not only Alex but with everyone on X) so that your posts are shown more often to him (and also to other users on X).

In this pursuit, your posts & profile is given a preference in this order:

[ >1 = advantage and <1 = disadvantage]

1. If you are Blue verified and are followed (by Alex), = 4x higher than a random unverified post author on X
2. If you are Blue verified and are not followed (by Alex) = still get a 2x higher than a random unverified post author on X.
3. If Alex favourites your post (Likes + Bookmarks) = 30x
4. If Alex reposts your post= 20x
5. If Alex follows you = 4x
6. If your posts carry an image & audio = 2x
7. If your post is in line with current trend = 1.1x
8. If you’ve used an unknown language/word in your post = (0.005) deboosted by 20x
9. If your post contains offensive words = (0.1) deboosted by 10x
10. If your post has multiple hashtags = (0.6) deboosted by 1.1x
11. If Alex has unfollowed you recently, muted, blocked, or reported you as spam = NO relationship is formed
12. If your post is NSFW, abusive, or toxic = NO relationship is formed

For a detailed metric along with relevant code snippets, refer to the detailed version.

From the relationship so formed, the top 1,500 posts are picked.

Ranking:

In this,

Notice here that the same “Data” is used to arrive at the probability values. However, X has not open-sourced how it arrives at these values depending on the data.

But the twist here is that all the 10 probabilities are not treated equally. Each carries a different weight.

[ +ve value = advantage and -ve value = disadvantage]

1. The probability that Alex will favourite your Post = 0.5
2. The probability that Alex will Retweet your Post = 1
3. The probability that Alex replies to your Post = 13.5
4. The probability that Alex opens your profile and Likes or Replies to a Post = 12
5. The probability (for a video Post) that Alex will watch at least half of the video = 0.005
6. The probability that Alex replies to the Post and this reply is engaged by you = 75
7. The probability that Alex will click into the conversation of this Post and reply or Like a Post = 11
8. The probability that Alex will click into the conversation of your Post and stay there for at least 2 minutes = 10
9. The probability that Alex will react negatively (requesting "show less often" on your Post or profile, block or mute you) = -74
10. The probability that Alex will click Report your Post = -369

Illustration: Let us assume you have made a post ‘ABC’ on X and the algorithm has assigned a probability value of 0.5 for the first 8 scenarios and a value of 0.001 for scenarios 9 and 10 based on {your, ABCs & Alex}’s Data.

💡 Remember that ABC’s Data and your Data is also used apart from Alex’s; so even if Alex had not reacted negatively to your profile or posts in the past, if your profile or ABC had been reported negatively by other users so far, it will negatively impact your chances of getting recommended to Alex as well.

Now, the score for the post ‘ABC’ will be:

(0.5 * 0.5) + (1 * 0.5) + (13.5 * 0.5) + (12 * 0.5) + (0.005 * 0.5) + (75 * 0.5) + (11 * 0.5) + (10 * 0.5) + (-74 * 0.001) + (-369 * 0.001) = 61.0595

For a detailed metric along with relevant code snippets, refer to the detailed version.

Based on the Feature Formation and ranking, a sorted list of 1,500 posts is made. They pass the filtering & heuristics and mixing stage before being finally displayed on Alex’s “For you” timeline.

Nerdy Nirvana: A Deep-Dive Into The Algorithm

Tools that X uses:

Before diving into the algorithm, it is vital to have an understanding of the following:

• Features: they act as inputs for all of X’s algorithm [they can be likes, reposts, replies, clicks, mutes, blocks, user data, etc.]
• Packages: These bots/programs take features as input and create a graph → cluster them → pick top posts from it → rank them → mix them with ads and serve the For you feed.
• Graphing: A relationship graph is created among various nodes (entities) within X. A sub-part of the process, graphs are the outcome of the machine-learning packages on X and it is based on this the 1,500 posts are picked.

A) Features:

For the purpose of the above question, X uses multiple features (categorized into 3 headers).

The data inputs such as Alex’s & Bella’s:

And many more are taken into consideration.

💡 There are 1000s of features on X. For a detailed list refer: Features.md

💡 The nodes in the above example, Alex & Bella, are users. In reality, they can be anything - from users, posts, clusters, media, and other things - making up billions of nodes.

B) Packages:

X has a gamut of packages (programs/bots). The most essential ones worth knowing are:

💡 Bonus: these packages are also used by X for other functionalities such as returning search results, follow-recommendation-service, etc.

c) Graphing:

Four key packages viz., RealGraph, GraphJet, TwHIN, and SimClusters get into work here and do the following:

1- Creates a graph between users.

Here, the users are nodes and the edges that connect them are the interactions (can be a like, a repost, etc.), and each edge is directed (A→B is different from B→A) and carries a weight (indicating relationship strength)

2- Creates a graph between users and posts

3- Associates users and posts into one cluster and creates a graph of clusters

To give a picture of how the features are used to create a graph, let us look at the Representation-Scorer package. It provides a scoring system for SimClusters

👉 For the Representation-Scorer:
{favourites, reposts, followings, shares, replies, posts, etc } = positive signals
{ blocks, mutes, reports, see-fewer } = negative signals

The Process:

Now that you've understood the core tech behind it, let’s put them to use. X breaks down the whole process into three sequential jobs.

1. Candidate Sourcing:

Trust-and-Safety works in parallel with all of the above.

2. Ranking:

Now that X has fetched 1,500 posts to be displayed, here comes the next challenge - in what order they are to be displayed to the user. X’s blog reads:

This is done by the HeavyRanker package and the ranking formula is:

score = sum_i { (weight of engagement i) * (probability of engagement i) }

There are 2 values here: weights and probability.

The probability can be between 0 & 1 → It is arrived by studying 1000s of features (refer to Feature section) of

1. The post itself, and
2. The user that the post is being recommended to

• Computing Probability:

Based on the 1000s of features, the model outputs 10 discreet labels and gives a value between 0 and 1 for each of the labels. The labels are:

1. The probability the user will favorite the Post.
2. The probability the user will Retweet the Post.
3. The probability the user replies to the Post.
4. The probability the user opens the Post author profile and Likes or replies to a Post.
5. The probability (for a video Post) that the user will watch at least half of the video.
6. The probability the user replies to the Post and this reply is engaged by the Post author.
7. The probability the user will click into the conversation of this Post and reply or Like a Post.
8. The probability the user will click into the conversation of this Post and stay there for at least 2 minutes.
9. The probability the user will react negatively (requesting "show less often" on the Post or author, block or mute the Post author).
10. The probability the user will click Report Post.

💡 PS: The algorithm on how X computes this probability based on the features (i.e., features → probability) has not been open-sourced yet, and is a point of major criticism.

• Weights: For each of the above labels, there is a pre-defined weight. This can be found in the Configuration File:

💡 Note: The weights are used to compute probability -i.e., they are assigned before such actions are taken on a post based on likelihood.

Based on the above formula [ Σ ( weight x probability ) ], a score is computed for the 1,500 tweets and ranked.

3. Filters, Heuristics & Product Features:

This is done to enhance product quality.

• Visibility Filtering: Hides posts from blocked or muted accounts. The visibility-filters package does this. It also filters out posts from accounts held back by legal agencies, suspended accounts, etc.
• Author Diversity: Prevent seeing too many posts from the same person in a row.
• Content Balance: Show a mix of posts from in-network and out-of-network
• Feedback-based Fatigue: Reduces visibility of posts that have received negative feedback from the user.
• Social Proof: Ensures out-of-network posts are shown only if some of the accounts followed by a user have interacted with them.
• Conversations: Provide more context to a Reply by threading it together with the original post.
• Edited Tweets: Keeps the feed fresh by replacing outdated tweets with their updated versions.

4. Mixing & Serving:

Ranked Posts + Ads + Follow-recommendations = "For you" feed

The HomeMixer package finally mixes the outcome of the above with Ads and Follower-recommendation-service (one that recommends whom to follow in the feed) placing them in-between tweets, and serves you an exclusively curated “For you”

Understanding Reach:

Based on the above process, there are four levels where you can take advantage as a post author to maximize reach.

While we take a look at each one of them, assume the perspective of a post-author

They are:

1. At the Candidate Sourcing Level

We’ve seen that RealGraph, GraphJet, TwHIN, and SimClusters try to make a graph of users, posts, clusters, etc. as nodes and establish a relationship between each node.

You & your posts are some of the many billion nodes on Twitter.

The code for determining this is:

private def getLinearRankingParams: ThriftRankingParams = {
    ThriftRankingParams(
      `type` = Some(ThriftScoringFunctionType.Linear),
      minScore = -1.0e100,
      retweetCountParams = Some(ThriftLinearFeatureRankingParams(weight = 20.0)),
      replyCountParams = Some(ThriftLinearFeatureRankingParams(weight = 1.0)),
      reputationParams = Some(ThriftLinearFeatureRankingParams(weight = 0.2)),
      luceneScoreParams = Some(ThriftLinearFeatureRankingParams(weight = 2.0)),
      textScoreParams = Some(ThriftLinearFeatureRankingParams(weight = 0.18)),
      urlParams = Some(ThriftLinearFeatureRankingParams(weight = 2.0)),
      isReplyParams = Some(ThriftLinearFeatureRankingParams(weight = 1.0)),
      favCountParams = Some(ThriftLinearFeatureRankingParams(weight = 30.0)),
      langEnglishUIBoost = 0.5,
      langEnglishTweetBoost = 0.2,
      langDefaultBoost = 0.02,
      unknownLanguageBoost = 0.05,
      offensiveBoost = 0.1,
      inTrustedCircleBoost = 3.0,
      multipleHashtagsOrTrendsBoost = 0.6,
      inDirectFollowBoost = 4.0,
      tweetHasTrendBoost = 1.1,
      selfTweetBoost = 2.0,
      tweetHasImageUrlBoost = 2.0,
      tweetHasVideoUrlBoost = 2.0,
      useUserLanguageInfo = true,
      ageDecayParams = Some(ThriftAgeDecayRankingParams(slope = 0.005, base = 1.0))

Source: GitHub

Since the graphing algorithms make use of data under three categories (Tweet Engagement, Follower Graph, and User Data), let us break the code into three parts and look at each one of them:

• Follower Graph: this is simpler, i.e., who follows you

NOTE: The TweepCred package impacted the credibility of users based on the followers-to-following ratio. It has since been deprecated.

• Tweet Engagement: this is the data about your posts.

There are also severe de-boosts:

• User Data: this is the data about you as a user.

While it collects information from a plethora of features like geolocation to business partner data, the essential ones are:

val allEdgeFeatures: SCollection[Edge] =
      getEdgeFeature(SCollection.unionAll(
					Seq(blocks, mutes, abuseReports, spamReports, unfollows)))

Source: GitHub

Getting blocked, muted, reported as abuse or spam, and being unfollowed hurts (for up to the next 90 days from any of them happening)

Unfollows are not as heavily penalized as the other 4

Also, if you are a Verified (Blue subscribed) user, you get a boost.

object BlueVerifiedAuthorInNetworkMultiplierParam
extends FSBoundedParam [Double] (
name = “home_mixer_blue_verified_author_in_network_multiplier",
default = 4.0,
min = 0.6,
max = 100.0
)
object BlueVerifiedAuthorOutOfNetworkMultiplierParam
extends FSBoundedParam [Double] (
name = “home_mixer_blue_verified_author_out_of_network_multiplier",
default = 2.0,
min = 0.6,
max = 100.0
)

Your posts get a minimum of 2x boost across X and for your followers, it gets a 4x boost.

• Escaping Trust & Safety Filters:

The exact keywords and topics in the filters are dynamic and keep changing over time. Until recently, posts on Ukraine were de-boosted.

Source: GitHub

As former head of Trust & Safety Yael Roth has said:

In short, there are 4 filters and you want to avoid each one of them:

pNSFWMedia: Model to detect tweets with NSFW images. This includes adult and porn content.
pNSFWText: Model to detect tweets with NSFW text, adult/sexual topics.
pToxicity: Model to detect toxic tweets. Toxicity includes marginal content like insults and certain types of harassment. Toxic content does not violate Twitter's terms of service.
pAbuse: Model to detect abusive content. This includes violations of Twitter's terms of service, including hate speech, targeted harassment and abusive behavior.

Source: GitHub

2. At the Ranking Level

a- Hacking the feature weight table

💡 not yet open-sourced by X

b- Hacking the probability weight table

The 10 probability scores as discussed above:

This is the key and most definitive aspect of the entire algorithm

scored_tweets_model_weight_fav: 0.5
scored_tweets_model_weight_retweet: 1.0
scored_tweets_model_weight_reply: 13.5
scored_tweets_model_weight_good_profile_click: 12.0
scored_tweets_model_weight_video_playback50: 0.005
scored_tweets_model_weight_reply_engaged_by_author: 75.0
scored_tweets_model_weight_good_click: 11.0
scored_tweets_model_weight_good_click_v2: 10.0
scored_tweets_model_weight_negative_feedback_v2: -74.0
scored_tweets_model_weight_report: -369.0

Source: GitHub and GitHub

💡 Note: The weights are used to compute probability -i.e., they are assigned before such actions are taken on a post based on likelihood.

3. At the Filtering, Heuristics, and Product Features Level

a- Timing your Posts:

Older posts are less relevant and are hence shown less often. Posts on X have a half-life of 360 minutes [6 hours] - This means that a post’s relevance will decrease by 50% every 6 hours

struct ThriftAgeDecayRankingParams {
// the rate in which the score of older tweets decreases
1: optional double slope = 6.003
// the age, in minutes, where the age score of a tweet is half of the latest tweet
2: optional double halflife = 360.0
// the minimal age decay score a tweet will have
3: optional double base = 0.6
}(persisted='true')

So, the first few engagements (likes, replies, and reposts) are critical. This factor has no hard & fast rule, and hence, experiment and time your posts when your target audiences are awake and are likely to be active on X.

b- Ensure that your posts do not fall in the extremities of the legal spectrum

At the HomeMixer Level [Optional]

Experiment with paid promotions and place your posts as Ads on X.

Actionable Intelligence

No-brainers

Getting Blue verified gives you an immediate boost of

1. 4x to your followers

object BlueVerifiedAuthorInNetworkMultiplierParam
extends FSBoundedParam [Double] (
name = “home_mixer_blue_verified_author_in_network_multiplier",
default = 4.0,
min = 0.6,
max = 100.0
)

1. and 2x across X

object BlueVerifiedAuthorOutOfNetworkMultiplierParam
extends FSBoundedParam [Double] (
name = “home_mixer_blue_verified_author_out_of_network_multiplier",
default = 2.0,
min = 0.6,
max = 100.0
)

Avoid using multiple hashtags.

1. You get immediately deboosted by 1.7x (i.e., by a factor of 0.6)

multipleHashtagsOrTrendsBoost = 0.6,

Avoid posting content that is NSFW (both media and text), abusive, or possibly misinformation (content that you are not sure of)

pNSFWMedia: Model to detect tweets with NSFW images. This includes adult and porn content.
pNSFWText: Model to detect tweets with NSFW text, adult/sexual topics.
pToxicity: Model to detect toxic tweets. Toxicity includes marginal content like insults and certain types of harassment. Toxic content does not violate Twitter's terms of service.
pAbuse: Model to detect abusive content. This includes violations of Twitter's terms of service, including hate speech, targeted harassment and abusive behavior.

Offensive words also get deboosted by 10x (i.e., by a factor of 0.1)

offensiveBoost = 0.1,

Avoid posting content on sensitive topics.

1. The exact keywords and topics in the filters are dynamic and keep changing over time. Until recently, posts on Ukraine were de-boosted. Check the GitHub page of the code to keep yourself updated on what X considers sensitive.

Stick to posting on one/a few topics. This way you develop a strong relationship with a cluster

1. The SimCluster package ensures that you have a strong relationship among members in a cluster. Source: SimCuster
2. Your post also gets a 3x boost to users in your Trusted Circle

inTrustedCircleBoost = 3.0,

1. Replying to content that is out-of-network also carries a penalty of 10. Unless you are sure that you can offset this by getting high engagement on such replies, try avoiding it

// subtractive penalty applied after boosts for out-of-network replies.
120: optional double OutOfNetworkReplyPenalty = 10.0

Quick-Wins

1. Try to use captive and relevant images and videos in your posts. They boost your post by 2x.

tweetHasImageUrlBoost = 2.0,
tweetHasVideoUrlBoost = 2.0,

1. Avoid making spelling errors or posting in unknown languages (you can try representing them through images instead). They get deboosted by 20x (i.e., by a factor of 0.05)

unknownLanguageBoost = 0.05,

1. Write long-form content or threads where the user might spend more than 2 minutes reading it

"recap.engagement.is_good_clicked_convo_desc_v2": The probability the user will click into the conversation of this Tweet and stay there for at least 2 minutes.
scored_tweets_model_weight_good_click_v2: 10.0

1. Reply to people who comment on your posts.

"recap.engagement.is_replied_reply_engaged_by_author": The probability the user replies to the Tweet and this reply is engaged by the Tweet author.
scored_tweets_model_weight_reply_engaged_by_author: 75.0

1. Sometimes, even having a captivating profile picture that might kindle a user to click your profile might help. If the user does so, ensure that they like your posts by pinning your top posts and by making sure your highlights tab is updated with your most interesting posts.

"recap.engagement.is_profile_clicked_and_profile_engaged": The probability the user opens the Tweet author profile and Likes or replies to a Tweet.
scored_tweets_model_weight_good_profile_click: 12.0

Challenging Choices

1. Writing good content: ultimately everything boils down to this. It is the central crux that holds all the other cards together. Every like, repost, comment, and bookmark that your posts get has a major boost (both at the Feature Formation stage and the Ranking stage).

retweetCountParams = Some(ThriftLinearFeatureRankingParams(weight = 20.0)),
replyCountParams = Some(ThriftLinearFeatureRankingParams(weight = 1.0)),
isReplyParams = Some(ThriftLinearFeatureRankingParams(weight = 1.0)),
favCountParams = Some(ThriftLinearFeatureRankingParams(weight = 30.0)),

"recap.engagement.is_favorited": The probability the user will favorite the Tweet.
"recap.engagement.is_favorited": 0.5
"recap.engagement.is_good_clicked_convo_desc_favorited_or_replied": The probability the user will click into the conversation of this Tweet and reply or Like a Tweet.
"recap.engagement.is_good_clicked_convo_desc_favorited_or_replied": 11*
"recap.engagement.is_replied": The probability the user replies to the Tweet.
"recap.engagement.is_replied": 27
"recap.engagement.is_retweeted": The probability the user will ReTweet the Tweet.
"recap.engagement.is_retweeted": 1

//parameters used by Representation-scorer
1: optional double fav1dLast10Max // max score from last 10 faves in the last 1 day
2: optional double fav1dLast10Avg // avg score from last 10 faves in the last 1 day
3: optional double fav7dLast10Max // max score from last 10 faves in the last 7 days
4: optional double fav7dLast10Avg // avg score from last 10 faves in the last 7 days
5: optional double retweet1dLast10Max // max score from last 10 retweets in the last 1 days
6: optional double retweet1dLast10Avg // avg score from last 10 retweets in the last 1 days
7: optional double retweet7dLast10Max // max score from last 10 retweets in the last 7 days
8: optional double retweet7dLast10Avg // avg score from last 10 retweets in the last 7 days
9: optional double follow7dLast10Max // max score from the last 10 follows in the last 7 days
10: optional double follow7dLast10Avg // avg score from the last 10 follows in the last 7 days
11: optional double follow30dLast10Max // max score from the last 10 follows in the last 30 days
12: optional double follow30dLast10Avg // avg score from the last 10 follows in the last 30 days
13: optional double share1dLast10Max // max score from last 10 shares in the last 1 day
14: optional double share1dLast10Avg // avg score from last 10 shares in the last 1 day
15: optional double share7dLast10Max // max score from last 10 shares in the last 7 days
16: optional double share7dLast10Avg // avg score from last 10 shares in the last 7 days
17: optional double reply1dLast10Max // max score from last 10 replies in the last 1 day
18: optional double reply1dLast10Avg // avg score from last 10 replies in the last 1 day
19: optional double reply7dLast10Max // max score from last 10 replies in the last 7 days
20: optional double reply7dLast10Avg // avg score from last 10 replies in the last 7 days

1. Write on topics that are in trend ( but limit the use of hashtags). Apart from the hashtags, also use keywords that are trending. Your content gets boosted by a factor of 1.1x

tweetHasTrendBoost = 1.1

1. Tweet at a time when your target audience is likely to be active on X. The half-life of content on X is 5 hours. This means that every 6 hours, the chance of your post being recommended to others is reduced by half. Experiment and arrive at an optimum time through trial & error.

struct ThriftAgeDecayRankingParams {
// the rate in which the score of older tweets decreases
1: optional double slope = 6.003
// the age, in minutes, where the age score of a tweet is half of the latest tweet
2: optional double halflife = 360.0
// the minimal age decay score a tweet will have
3: optional double base = 0.6
}(persisted='true')

Avoiding Pitfalls

Creating posts with unconventional opinions, that run counter to public acceptance may attract significant attention in terms of clicks and replies. However, it can also result in numerous reports, mutes, and unfollows. Avoid getting muted, reported, unfollowed, and blocked.

val allEdgeFeatures: SCollection[Edge] =
      getEdgeFeature(SCollection.unionAll(
					Seq(blocks, mutes, abuseReports, spamReports, unfollows)))

val allEdgeFeatures: SCollection[Edge] =
      getEdgeFeature(SCollection.unionAll(Seq(blocks, mutes, abuseReports, spamReports, unfollows)))

    val negativeFeatures: SCollection[KeyVal[Long, UserSession]] =
      allEdgeFeatures
        .keyBy(_.sourceId)
        .topByKey(maxDestinationIds)(Ordering.by(_.features.size))
        .map {
          case (srcId, pqEdges) =>
            val topKNeg =
              pqEdges.toSeq.flatMap(toRealGraphEdgeFeatures(hasNegativeFeatures))
            KeyVal(
              srcId,
              UserSession(
                userId = Some(srcId),
                realGraphFeaturesTest =
                  Some(RealGraphFeaturesTest.V1(RealGraphFeaturesV1(topKNeg)))))

//parameters from representation scorer
// 2001 - 3000 Negative Signals
  // Block Series
  2001: optional double block1dLast10Avg
  2002: optional double block1dLast10Max
  2003: optional double block7dLast10Avg
  2004: optional double block7dLast10Max
  2005: optional double block30dLast10Avg
  2006: optional double block30dLast10Max
  // Mute Series
  2101: optional double mute1dLast10Avg
  2102: optional double mute1dLast10Max
  2103: optional double mute7dLast10Avg
  2104: optional double mute7dLast10Max
  2105: optional double mute30dLast10Avg
  2106: optional double mute30dLast10Max
  // Report Series
  2201: optional double report1dLast10Avg
  2202: optional double report1dLast10Max
  2203: optional double report7dLast10Avg
  2204: optional double report7dLast10Max
  2205: optional double report30dLast10Avg
  2206: optional double report30dLast10Max
  // Dontlike
  2301: optional double dontlike1dLast10Avg
  2302: optional double dontlike1dLast10Max
  2303: optional double dontlike7dLast10Avg
  2304: optional double dontlike7dLast10Max
  2305: optional double dontlike30dLast10Avg
  2306: optional double dontlike30dLast10Max
  // SeeFewer
  2401: optional double seeFewer1dLast10Avg
  2402: optional double seeFewer1dLast10Max
  2403: optional double seeFewer7dLast10Avg
  2404: optional double seeFewer7dLast10Max
  2405: optional double seeFewer30dLast10Avg
  2406: optional double seeFewer30dLast10Max