{
  "total": 330,
  "topics": [
    {
      "id": 363973,
      "title": "Welcome to the OTTO – RecSys Challenge 2022!",
      "comment_count": 62,
      "views": 0,
      "votes": 55
    },
    {
      "id": 411517,
      "title": "Complete OTTO Session Dataset Now Available!",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 384844,
      "title": "Update on Competition Compliance",
      "comment_count": 21,
      "views": 0,
      "votes": 68
    },
    {
      "id": 383133,
      "title": "Thanks to all the participants!",
      "comment_count": 8,
      "views": 0,
      "votes": 63
    },
    {
      "id": 363496,
      "title": "New to Kaggle or Machine Learning? Check this out ~",
      "comment_count": 3,
      "views": 0,
      "votes": 16
    },
    {
      "id": 363497,
      "title": "Looking for a Team Megathread",
      "comment_count": 153,
      "views": 0,
      "votes": 22
    },
    {
      "id": 384852,
      "title": "Poteman and RunningZ removed from Kaggle",
      "comment_count": 21,
      "views": 0,
      "votes": 49
    },
    {
      "id": 383013,
      "title": "3rd Place - Using Only Rules Achieves LB 0.590!",
      "comment_count": 55,
      "views": 0,
      "votes": 151
    },
    {
      "id": 368367,
      "title": "Has anyone tried PySpark?",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 384120,
      "title": "6th place solution (single model LB 0.603)",
      "comment_count": 10,
      "views": 0,
      "votes": 44
    },
    {
      "id": 383130,
      "title": "9th Place Solution",
      "comment_count": 6,
      "views": 0,
      "votes": 43
    },
    {
      "id": 384022,
      "title": "1st place solution",
      "comment_count": 41,
      "views": 0,
      "votes": 207
    },
    {
      "id": 370210,
      "title": "How To Build a GBT Ranker Model",
      "comment_count": 224,
      "views": 0,
      "votes": 335
    },
    {
      "id": 363862,
      "title": "What does the image from the evaluation section mean?!",
      "comment_count": 3,
      "views": 0,
      "votes": 9
    },
    {
      "id": 381318,
      "title": "Abnormal scores of our team",
      "comment_count": 33,
      "views": 0,
      "votes": 133
    },
    {
      "id": 364721,
      "title": "Recommendation Systems for Large Datasets",
      "comment_count": 39,
      "views": 0,
      "votes": 275
    },
    {
      "id": 373164,
      "title": "The Recommendation systems reading list you should have",
      "comment_count": 3,
      "views": 0,
      "votes": 33
    },
    {
      "id": 363587,
      "title": "❓📚💥 What is a Session - Some great resources for eCommerce Sessions💥📚❓",
      "comment_count": 2,
      "views": 0,
      "votes": 10
    },
    {
      "id": 413566,
      "title": "Test_labels data",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 401487,
      "title": "Co-visitation Matrix in other way",
      "comment_count": 0,
      "views": 0,
      "votes": -1
    },
    {
      "id": 386497,
      "title": "3rd place - Benny's part - XGB Reranker with Transformers+GRU input - LB0.601",
      "comment_count": 2,
      "views": 0,
      "votes": 18
    },
    {
      "id": 382790,
      "title": "2nd Place Solution(ONODERA part)",
      "comment_count": 34,
      "views": 0,
      "votes": 90
    },
    {
      "id": 373257,
      "title": "PySpark: How to efficiently read the data & write as parquet",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 382771,
      "title": "20th Place Solution",
      "comment_count": 31,
      "views": 0,
      "votes": 90
    },
    {
      "id": 384636,
      "title": "3rd Place Code ",
      "comment_count": 4,
      "views": 0,
      "votes": 28
    },
    {
      "id": 382975,
      "title": "3rd Place Solution - Theo's Part",
      "comment_count": 2,
      "views": 0,
      "votes": 49
    },
    {
      "id": 383121,
      "title": "17th place solution",
      "comment_count": 1,
      "views": 0,
      "votes": 13
    },
    {
      "id": 384847,
      "title": "We got a new GM after all!",
      "comment_count": 5,
      "views": 0,
      "votes": 16
    },
    {
      "id": 384482,
      "title": "So cheater GrandMaster Survive? Great job",
      "comment_count": 37,
      "views": 0,
      "votes": 48
    },
    {
      "id": 388747,
      "title": "Why covisitation instead of SVD?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 383016,
      "title": "46th Place Solution (Code and Journey)",
      "comment_count": 5,
      "views": 0,
      "votes": 23
    },
    {
      "id": 384441,
      "title": "13th place.  BST transformer model",
      "comment_count": 10,
      "views": 0,
      "votes": 16
    },
    {
      "id": 383769,
      "title": "7th Place Solution",
      "comment_count": 11,
      "views": 0,
      "votes": 84
    },
    {
      "id": 382879,
      "title": "3rd place(imaginary) solution",
      "comment_count": 20,
      "views": 0,
      "votes": 98
    },
    {
      "id": 383566,
      "title": "Be careful",
      "comment_count": 25,
      "views": 0,
      "votes": 42
    },
    {
      "id": 384881,
      "title": "JUSTICE HAS BEEN DONE",
      "comment_count": 8,
      "views": 0,
      "votes": 12
    },
    {
      "id": 384562,
      "title": "Another Coincidence: Abnormally High Similarity of Submissions",
      "comment_count": 8,
      "views": 0,
      "votes": 35
    },
    {
      "id": 384483,
      "title": "Congratulations to new Kaggle Competition GM",
      "comment_count": 10,
      "views": 0,
      "votes": 16
    },
    {
      "id": 382802,
      "title": "5th place Solution",
      "comment_count": 32,
      "views": 0,
      "votes": 103
    },
    {
      "id": 381321,
      "title": "About my team's result and activities",
      "comment_count": 80,
      "views": 0,
      "votes": 128
    },
    {
      "id": 384613,
      "title": "What did this competition really bring to us?",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 383792,
      "title": "9th place solution🥇 [single model LB 0.602]",
      "comment_count": 2,
      "views": 0,
      "votes": 43
    },
    {
      "id": 384598,
      "title": "Small step to session-based recommend system.",
      "comment_count": 0,
      "views": 0,
      "votes": 7
    },
    {
      "id": 382776,
      "title": "ex-27th place solution",
      "comment_count": 15,
      "views": 0,
      "votes": 34
    },
    {
      "id": 383229,
      "title": "13th Place Solution",
      "comment_count": 5,
      "views": 0,
      "votes": 27
    },
    {
      "id": 382834,
      "title": "12th place solution",
      "comment_count": 17,
      "views": 0,
      "votes": 37
    },
    {
      "id": 383382,
      "title": "16th Place Solution (Association x UserCF x NN-based x Matrix Factorization x Covisit x LightGBM)",
      "comment_count": 2,
      "views": 0,
      "votes": 28
    },
    {
      "id": 382839,
      "title": "2nd Place Solution(senkin13&30CrMnSiA part)",
      "comment_count": 16,
      "views": 0,
      "votes": 69
    },
    {
      "id": 383843,
      "title": "34th (ex 37th) Place Solution (Polars is here to stay !)",
      "comment_count": 1,
      "views": 0,
      "votes": 13
    },
    {
      "id": 384075,
      "title": "207th using only a model for Carts",
      "comment_count": 1,
      "views": 0,
      "votes": 9
    },
    {
      "id": 383374,
      "title": "14th Place Solution(Ethan&qyxs part)",
      "comment_count": 3,
      "views": 0,
      "votes": 23
    },
    {
      "id": 383657,
      "title": "226th (?!) Place Solution & Two-cents from a First-timer",
      "comment_count": 2,
      "views": 0,
      "votes": 12
    },
    {
      "id": 382782,
      "title": "Organized cheating again",
      "comment_count": 15,
      "views": 0,
      "votes": 41
    },
    {
      "id": 383493,
      "title": "17th place solution(dehokanta-part)",
      "comment_count": 0,
      "views": 0,
      "votes": 15
    },
    {
      "id": 382812,
      "title": "28th Place Solution",
      "comment_count": 4,
      "views": 0,
      "votes": 34
    },
    {
      "id": 382955,
      "title": "Idea about the \"Cross Target Stacking\" Method",
      "comment_count": 7,
      "views": 0,
      "votes": 22
    },
    {
      "id": 382851,
      "title": "10th Place Solution",
      "comment_count": 13,
      "views": 0,
      "votes": 31
    },
    {
      "id": 382924,
      "title": "20th place solution (Transformer inside)",
      "comment_count": 4,
      "views": 0,
      "votes": 29
    },
    {
      "id": 382772,
      "title": "116th place write up",
      "comment_count": 6,
      "views": 0,
      "votes": 36
    },
    {
      "id": 382783,
      "title": "5th place (yet) solution (Carno & 2U & Jiahong's part)",
      "comment_count": 5,
      "views": 0,
      "votes": 38
    },
    {
      "id": 382841,
      "title": "This was a really good competition",
      "comment_count": 5,
      "views": 0,
      "votes": 28
    },
    {
      "id": 381360,
      "title": "One more thread about cheaters",
      "comment_count": 47,
      "views": 0,
      "votes": 84
    },
    {
      "id": 382905,
      "title": "15th Place Solution",
      "comment_count": 6,
      "views": 0,
      "votes": 19
    },
    {
      "id": 381527,
      "title": "How cheaters can steal tiers and reputations?",
      "comment_count": 24,
      "views": 0,
      "votes": 87
    },
    {
      "id": 382935,
      "title": "5th place (yet) solution (@EEHITer & @wj19971997)",
      "comment_count": 0,
      "views": 0,
      "votes": 16
    },
    {
      "id": 382440,
      "title": "Should we be careful about sharing our solutions after the contest?",
      "comment_count": 12,
      "views": 0,
      "votes": 7
    },
    {
      "id": 382910,
      "title": "168th place solution(Bronze medal)",
      "comment_count": 4,
      "views": 0,
      "votes": 9
    },
    {
      "id": 382829,
      "title": "70-th Place Solution : Everything done on Kaggle",
      "comment_count": 0,
      "views": 0,
      "votes": 17
    },
    {
      "id": 382886,
      "title": "17th place solution (T88 part)",
      "comment_count": 0,
      "views": 0,
      "votes": 14
    },
    {
      "id": 381588,
      "title": "Connecting more dots: private sharing & medal selling by top competitors in this competition.",
      "comment_count": 12,
      "views": 0,
      "votes": 55
    },
    {
      "id": 382884,
      "title": "[Silver] 113th - Computationally simple solution",
      "comment_count": 4,
      "views": 0,
      "votes": 9
    },
    {
      "id": 382911,
      "title": "How do I add my solution in the leaderboard tab ?",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 381495,
      "title": "【Maybe massive cheating】 Weird LB changes around silver",
      "comment_count": 28,
      "views": 0,
      "votes": 24
    },
    {
      "id": 382774,
      "title": "Maybe, just maybe, let's not share code until competition results verified?",
      "comment_count": 3,
      "views": 0,
      "votes": 12
    },
    {
      "id": 382277,
      "title": "The mistake that made me lose 1 month in this competition",
      "comment_count": 11,
      "views": 0,
      "votes": 18
    },
    {
      "id": 383260,
      "title": "Proof of concept vanilla \"Markov Decision Process\"",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 382808,
      "title": "Thanks Otto",
      "comment_count": 1,
      "views": 0,
      "votes": 9
    },
    {
      "id": 382897,
      "title": "69th story and solution",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 382921,
      "title": "110th place solution",
      "comment_count": 0,
      "views": 0,
      "votes": 7
    },
    {
      "id": 382542,
      "title": "Guess the team will win this challenge!",
      "comment_count": 6,
      "views": 0,
      "votes": 5
    },
    {
      "id": 382909,
      "title": "Looking forward a medal winning solution using deep learning",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 380783,
      "title": "Be careful of deadline teamup",
      "comment_count": 12,
      "views": 0,
      "votes": 35
    },
    {
      "id": 379631,
      "title": "Training GBDT Ranker based on Chris's discussion",
      "comment_count": 18,
      "views": 0,
      "votes": 23
    },
    {
      "id": 382836,
      "title": "so much fun",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 381717,
      "title": "Questions for the TOP 100 : Are you using only ranker  ?",
      "comment_count": 10,
      "views": 0,
      "votes": 4
    },
    {
      "id": 382254,
      "title": "Discussion: If Collaborative Filtering Algorithm is suitable for this challenge or not?",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 381787,
      "title": "What is your next target competition",
      "comment_count": 6,
      "views": 0,
      "votes": 6
    },
    {
      "id": 381471,
      "title": "LB Rank Analysis using Kaggle meta-data",
      "comment_count": 3,
      "views": 0,
      "votes": 18
    },
    {
      "id": 382286,
      "title": "Hash trick to save Memory when merging data (64GB for all)",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 377623,
      "title": "A good score with NN as reranker?",
      "comment_count": 9,
      "views": 0,
      "votes": 7
    },
    {
      "id": 382103,
      "title": "Question about the interactions features",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 381469,
      "title": "Top 200 : Empirical results/remarks about training the ranker.",
      "comment_count": 9,
      "views": 0,
      "votes": 7
    },
    {
      "id": 382543,
      "title": "Last 24 hours final submission select suggest.",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 382031,
      "title": "How  we get First position ranker notebook required after competition over",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 381630,
      "title": "Can we make the miracle with larger computation resources?",
      "comment_count": 9,
      "views": 0,
      "votes": 6
    },
    {
      "id": 381773,
      "title": "Why LGBM ndcg nearly 0.9 while LB score is only 0.4?",
      "comment_count": 9,
      "views": 0,
      "votes": 3
    },
    {
      "id": 381672,
      "title": "X15 Faster Inference : Using the TPU VM -3-8 vs CPU",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 365358,
      "title": "💡 What is the co-visitation matrix, really?",
      "comment_count": 18,
      "views": 0,
      "votes": 66
    },
    {
      "id": 379605,
      "title": "Obvious Tips ",
      "comment_count": 6,
      "views": 0,
      "votes": 6
    },
    {
      "id": 382289,
      "title": "How much did the ensemble improve LB?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 382048,
      "title": "What is your best negative sampling rate ？ ",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 382051,
      "title": "using all data to train model without validating.   vs    KFold&Ensemble",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 380772,
      "title": "Report Cheaters!",
      "comment_count": 0,
      "views": 0,
      "votes": 21
    },
    {
      "id": 381587,
      "title": "Is there any ways to automate sequential execution of notebooks ?",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 379952,
      "title": "Some optimization tips for training ranker model",
      "comment_count": 7,
      "views": 0,
      "votes": 31
    },
    {
      "id": 380169,
      "title": "Last tips that helps me beat the benchmark recall using GBDT",
      "comment_count": 2,
      "views": 0,
      "votes": 16
    },
    {
      "id": 379860,
      "title": "I guess it's time for the pre-deadline meme thread",
      "comment_count": 5,
      "views": 0,
      "votes": 22
    },
    {
      "id": 379741,
      "title": "How much did you get without interactions features?",
      "comment_count": 13,
      "views": 0,
      "votes": 5
    },
    {
      "id": 381501,
      "title": "What happens if the ground truth has no carts or orders after truncation point?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 380732,
      "title": "Similarities between users and candidates",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 380379,
      "title": "I can't use early stopping for LGBMranker.",
      "comment_count": 6,
      "views": 0,
      "votes": 3
    },
    {
      "id": 379980,
      "title": "My GBDT Ranker can't beat heuristic approach !?",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 380126,
      "title": "If chatGPT was a Kaggle player",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 375297,
      "title": "Increase your RAM to 330GB on kaggle.",
      "comment_count": 26,
      "views": 0,
      "votes": 54
    },
    {
      "id": 378893,
      "title": "Session (User) features are not effective in my GBM model.",
      "comment_count": 18,
      "views": 0,
      "votes": 2
    },
    {
      "id": 379291,
      "title": "A growing reflections on otto from a beginner's perspective",
      "comment_count": 3,
      "views": 0,
      "votes": 6
    },
    {
      "id": 378947,
      "title": "Congratulations to Radek on becoming Kaggle Grandmaster 🙏",
      "comment_count": 10,
      "views": 0,
      "votes": 27
    },
    {
      "id": 377442,
      "title": "NDCG@20 = 0.99 after re-ranking orders candidates",
      "comment_count": 14,
      "views": 0,
      "votes": 2
    },
    {
      "id": 380352,
      "title": "Time limit on training?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 376789,
      "title": "How much of your max-recall do you get?",
      "comment_count": 16,
      "views": 0,
      "votes": 21
    },
    {
      "id": 376693,
      "title": "Do you expect better recall after using a Ranker ?",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 377894,
      "title": "Revealing the 4th digit of the score",
      "comment_count": 14,
      "views": 0,
      "votes": 13
    },
    {
      "id": 377697,
      "title": "Question for the top 100 of the competition",
      "comment_count": 12,
      "views": 0,
      "votes": 4
    },
    {
      "id": 379177,
      "title": "Something goes wrong on my local cv",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 379431,
      "title": "Organizing my growing notebooks for otto",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 378421,
      "title": "[Chai Time] H&M Top Solutions Explained",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 374824,
      "title": "Why my rank model doesn't perform better than the public notebook",
      "comment_count": 5,
      "views": 0,
      "votes": 4
    },
    {
      "id": 378058,
      "title": "classification v.s. rank which is better objective for the reranking model",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 378786,
      "title": "Submission",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 377511,
      "title": "[Video] Detailed Walkthrough & Overview of the Comp by Radek",
      "comment_count": 3,
      "views": 0,
      "votes": 15
    },
    {
      "id": 378151,
      "title": "The way to split test dataset into public/private",
      "comment_count": 5,
      "views": 0,
      "votes": 5
    },
    {
      "id": 378219,
      "title": "Generate 1-week or any length of `train-test-test_label` dataset for fast iteration on Kaggle",
      "comment_count": 0,
      "views": 0,
      "votes": 9
    },
    {
      "id": 370502,
      "title": "Ranker models vs. Binary classification models",
      "comment_count": 5,
      "views": 0,
      "votes": 15
    },
    {
      "id": 370751,
      "title": "I think word2vec is a nice choice.",
      "comment_count": 20,
      "views": 0,
      "votes": 40
    },
    {
      "id": 378454,
      "title": "Do we need to use different co-visitation matrix for train and inference ?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 378604,
      "title": "Dumb question : which LightGBMRanker.predict() doesn't give a prediction between 0 and 1 ?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 374229,
      "title": "One Month Left - Here is what you need to know!",
      "comment_count": 8,
      "views": 0,
      "votes": 89
    },
    {
      "id": 373445,
      "title": "Which category are you best at ?",
      "comment_count": 9,
      "views": 0,
      "votes": 9
    },
    {
      "id": 370756,
      "title": "Make NN Great Again!!",
      "comment_count": 19,
      "views": 0,
      "votes": 41
    },
    {
      "id": 377632,
      "title": "Question on Neural Collaborative Filtering",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 377149,
      "title": "Big wall in LB: only 4.12% of teams scored >= 0.579",
      "comment_count": 2,
      "views": 0,
      "votes": 9
    },
    {
      "id": 377144,
      "title": "Nothing Useful Just Talking",
      "comment_count": 5,
      "views": 0,
      "votes": 10
    },
    {
      "id": 377395,
      "title": "How to immpot .jsonl dataset?",
      "comment_count": 8,
      "views": 0,
      "votes": null
    },
    {
      "id": 377094,
      "title": "Tips: leakage in ranking model: the order of dataset",
      "comment_count": 8,
      "views": 0,
      "votes": 8
    },
    {
      "id": 376977,
      "title": "Journal Papers about Recommender Systems",
      "comment_count": 1,
      "views": 0,
      "votes": 12
    },
    {
      "id": 377608,
      "title": "May I ask how to prepar labels for a re-rank model ??",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 377324,
      "title": "Candidate rerank model recommends less than 20 aids per session",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 376905,
      "title": "Is it necessary for us to write an recall@20 metric for the competition",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 376573,
      "title": "Candidates evaluation tricks from previous winning solutions",
      "comment_count": 0,
      "views": 0,
      "votes": 13
    },
    {
      "id": 372030,
      "title": "What is wrong with my GBT Ranker Pipeline? [LB = CV -> lower than expected]",
      "comment_count": 31,
      "views": 0,
      "votes": 19
    },
    {
      "id": 368000,
      "title": "💡 how to move faster on an ML project and achieve more with less compute/time/energy (relevant to this competition)",
      "comment_count": 14,
      "views": 0,
      "votes": 31
    },
    {
      "id": 375581,
      "title": "How many rounds/epochs do you train ?",
      "comment_count": 23,
      "views": 0,
      "votes": 15
    },
    {
      "id": 368170,
      "title": "💡 How to deal with this competition needing so much RAM -- a couple of things that worked for me",
      "comment_count": 12,
      "views": 0,
      "votes": 28
    },
    {
      "id": 376994,
      "title": "NDCG@20 increases while loss increases",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 376547,
      "title": "Question about the test data",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 376191,
      "title": "How to fix \"Duplicate bug\" in current baseline notebooks",
      "comment_count": 6,
      "views": 0,
      "votes": 8
    },
    {
      "id": 376443,
      "title": "4x Speedup Pandas using [Pandarallel and TPU]",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 370116,
      "title": "Surprising LB 0.587 !!!Share some experimental results ~~",
      "comment_count": 35,
      "views": 0,
      "votes": 112
    },
    {
      "id": 374677,
      "title": "Some experimental approaches",
      "comment_count": 7,
      "views": 0,
      "votes": 13
    },
    {
      "id": 375008,
      "title": " Multi-Gradient Descent approach",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 376563,
      "title": "Question about final test set and evaluation",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 375820,
      "title": "Association Rule Mining is all we need?",
      "comment_count": 3,
      "views": 0,
      "votes": 4
    },
    {
      "id": 376120,
      "title": "Can we use Word2Vec ? ",
      "comment_count": 1,
      "views": 0,
      "votes": 0
    },
    {
      "id": 376026,
      "title": "Is it a rule violation?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 374324,
      "title": "How many candidates did you generates and how did they perform?",
      "comment_count": 19,
      "views": 0,
      "votes": 13
    },
    {
      "id": 375224,
      "title": " We're just Rebuilding Otto's Recommendation System?",
      "comment_count": 9,
      "views": 0,
      "votes": 14
    },
    {
      "id": 368564,
      "title": "Evaluation of stage1.Candidates",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 375925,
      "title": "Should I setup the field `type` of the candidates not in the user behavior history to a new value, like 3 ? ",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 375192,
      "title": "💡 How to ensemble predictions and improve your score?",
      "comment_count": 0,
      "views": 0,
      "votes": 10
    },
    {
      "id": 364991,
      "title": "local validation tracks public LB perfecty -- here is the setup",
      "comment_count": 55,
      "views": 0,
      "votes": 208
    },
    {
      "id": 375148,
      "title": "can we use session id find the sessions of the same user in the training session?",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 371058,
      "title": "Saving GPU memory when processing features == (Speedup + Efficiency)",
      "comment_count": 4,
      "views": 0,
      "votes": 42
    },
    {
      "id": 374994,
      "title": "Can we use  KNN or K means clustering algorithm to generate recommendation?",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 375178,
      "title": "From 0.575 to 0.000 :)",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 374902,
      "title": "Reduce Otto Training Set",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 374493,
      "title": "Create your own validation set with organizer's script on Kaggle",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 368560,
      "title": "💡 For my friends from Twitter and LinkedIn -- here is how to dive into this competition 🐳",
      "comment_count": 8,
      "views": 0,
      "votes": 136
    },
    {
      "id": 373224,
      "title": "Otto's Talk on Transformer Recommendation Systems",
      "comment_count": 2,
      "views": 0,
      "votes": 24
    },
    {
      "id": 371678,
      "title": "The Story of User 13479136 and User 13710374",
      "comment_count": 15,
      "views": 0,
      "votes": 79
    },
    {
      "id": 374785,
      "title": "The use of time stamp in the project",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 368278,
      "title": "📑 [Step-by-step guide] How I got to my current standing on the LB and how to improve going forward",
      "comment_count": 20,
      "views": 0,
      "votes": 78
    },
    {
      "id": 374538,
      "title": "Question about kaggle notebooks for beginer",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 373583,
      "title": "Not enough memory in Kaggle notebook!",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 367154,
      "title": "Dumb Question: so what do we actually need to predict?",
      "comment_count": 15,
      "views": 0,
      "votes": 34
    },
    {
      "id": 373722,
      "title": "Do we really need to use cross-validation over holdout in large datasets to win a competition?",
      "comment_count": 5,
      "views": 0,
      "votes": 3
    },
    {
      "id": 374502,
      "title": "My LGBMRanker performs poorly ",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 372781,
      "title": "Kaggle's Top Recommender Systems Notebooks",
      "comment_count": 1,
      "views": 0,
      "votes": 23
    },
    {
      "id": 372927,
      "title": "Visualizing 3D and 4D Embeddings",
      "comment_count": 1,
      "views": 0,
      "votes": 17
    },
    {
      "id": 374192,
      "title": "Has anyone encountered a MemoryError like this?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 371916,
      "title": "Duplicates in train dataset",
      "comment_count": 2,
      "views": 0,
      "votes": 9
    },
    {
      "id": 373403,
      "title": "Negative Sampling of Candidates",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 368355,
      "title": "Graph Neural Netwok , is complex to model but efficient for this task ",
      "comment_count": 9,
      "views": 0,
      "votes": 15
    },
    {
      "id": 371991,
      "title": "Top Ranking Teams what's your machine spec ?",
      "comment_count": 3,
      "views": 0,
      "votes": 7
    },
    {
      "id": 372976,
      "title": "Winning solutions summary from H&M Recommendations comp",
      "comment_count": 1,
      "views": 0,
      "votes": 40
    },
    {
      "id": 370950,
      "title": "💡 Implicit vs Explicit data -- a useful distinction for recommender systems",
      "comment_count": 3,
      "views": 0,
      "votes": 11
    },
    {
      "id": 372157,
      "title": "a question about the session",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 372990,
      "title": "How did you read the data?",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 368618,
      "title": "How much RAM do I need for this comp?",
      "comment_count": 7,
      "views": 0,
      "votes": 11
    },
    {
      "id": 373163,
      "title": "Exposure Bias and Its Impact on Recommender Systems",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 372945,
      "title": "GPU Inference by RAPIDS",
      "comment_count": 0,
      "views": 0,
      "votes": 8
    },
    {
      "id": 368240,
      "title": "Some learning materials I found",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 363843,
      "title": "Full dataset processed to CSV/parquet files with optimized memory footprint",
      "comment_count": 28,
      "views": 0,
      "votes": 77
    },
    {
      "id": 372353,
      "title": "I want to know recommendation systems Learning resources?",
      "comment_count": 12,
      "views": 0,
      "votes": 6
    },
    {
      "id": 372585,
      "title": "Difference between Co-visitation Matrix v.s. Matrix Factorization",
      "comment_count": 0,
      "views": 0,
      "votes": 11
    },
    {
      "id": 372552,
      "title": "p(cart, click) = p(cart | click) p(click)",
      "comment_count": 0,
      "views": 0,
      "votes": 10
    },
    {
      "id": 372521,
      "title": "Heuristic to find promoted items",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 372289,
      "title": "Test data generation",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 365899,
      "title": "What is the output we are expected to give?",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 372452,
      "title": "Resources to Get Started with this competition",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 369789,
      "title": "To the people still hacking away on this competition...",
      "comment_count": 14,
      "views": 0,
      "votes": 20
    },
    {
      "id": 371910,
      "title": "About the Logic of Aid?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 368538,
      "title": "3 dimensions Co-visitation matrix",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 371731,
      "title": "Very confused in understanding the candidates generation and feature-engineering",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 367234,
      "title": "How to train a Word2Vec model for item embeddings - a simple code example 📖",
      "comment_count": 9,
      "views": 0,
      "votes": 43
    },
    {
      "id": 371166,
      "title": "Matrix Factorization with GPU: 6.5x faster!",
      "comment_count": 4,
      "views": 0,
      "votes": 26
    },
    {
      "id": 371111,
      "title": "💡 ANN -> NN: get better results and run faster with NN search on the GPU 🔥🔥🔥",
      "comment_count": 0,
      "views": 0,
      "votes": 19
    },
    {
      "id": 368373,
      "title": "Some Interesting Times Series on Products",
      "comment_count": 2,
      "views": 0,
      "votes": 13
    },
    {
      "id": 371476,
      "title": "Fast Recall@20",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 367573,
      "title": "Dumb question: So what's the practical use-case of such a model?",
      "comment_count": 6,
      "views": 0,
      "votes": 7
    },
    {
      "id": 367991,
      "title": "The time zone in Germany is UTC+2 in August",
      "comment_count": 5,
      "views": 0,
      "votes": 18
    },
    {
      "id": 368685,
      "title": "💡What is a good initial goal in the competition? How to improve beyond it? 📈",
      "comment_count": 2,
      "views": 0,
      "votes": 23
    },
    {
      "id": 363939,
      "title": "Yes, We Can Use Test Data Leakage.",
      "comment_count": 17,
      "views": 0,
      "votes": 51
    },
    {
      "id": 371266,
      "title": "How to correctly filter data to process in kaggle notebooks (30 GB ram)",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 370813,
      "title": "Can we use fastai for building a recommendation system for OTTO?",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 371439,
      "title": "[revisit] the problem",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 365873,
      "title": "Say NO to GPU: Nx Faster Co-Visitation Matrices using single CPU!",
      "comment_count": 9,
      "views": 0,
      "votes": 43
    },
    {
      "id": 370384,
      "title": "Some concerns about validation",
      "comment_count": 2,
      "views": 0,
      "votes": 10
    },
    {
      "id": 370070,
      "title": "How can a session start with an order?",
      "comment_count": 5,
      "views": 0,
      "votes": 10
    },
    {
      "id": 370404,
      "title": "Minimum /Maximum History data ",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 370347,
      "title": "Looking for a teamate, I'm a ML engineer based in Guangzhou",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 369680,
      "title": "What is your best score of pure co-visit-mat method?",
      "comment_count": 8,
      "views": 0,
      "votes": 16
    },
    {
      "id": 368848,
      "title": "💡 Training an XGBoost Ranker on the GPU with Merlin Models 🔥🔥🔥",
      "comment_count": 2,
      "views": 0,
      "votes": 22
    },
    {
      "id": 369728,
      "title": "How many candidates you used?",
      "comment_count": 6,
      "views": 0,
      "votes": 6
    },
    {
      "id": 367503,
      "title": "How to thrive in this competition without going crazy -- 1 out of 2 important truths ❤️‍🔥",
      "comment_count": 7,
      "views": 0,
      "votes": 61
    },
    {
      "id": 369464,
      "title": "💡 How to add generated candidates to your train data",
      "comment_count": 1,
      "views": 0,
      "votes": 9
    },
    {
      "id": 370091,
      "title": "problem, how to solve it?   the data is too big",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 366138,
      "title": "Locate Real Users and Real Sessions EDA",
      "comment_count": 12,
      "views": 0,
      "votes": 80
    },
    {
      "id": 365369,
      "title": "30x Faster Co-Visitation Matrices using RAPIDS cuDF!",
      "comment_count": 32,
      "views": 0,
      "votes": 88
    },
    {
      "id": 369419,
      "title": "💡 How to split the data for training a two-stage recommender?",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 369772,
      "title": "Is the timestamp sorted?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 363965,
      "title": "Important information regarding test data from competition repository",
      "comment_count": 12,
      "views": 0,
      "votes": 21
    },
    {
      "id": 366709,
      "title": "Summary about Loading and Preprocessing Big Jsonl Data File",
      "comment_count": 2,
      "views": 0,
      "votes": 12
    },
    {
      "id": 368885,
      "title": "Transformer + cls head is too slow, Need new train framework",
      "comment_count": 5,
      "views": 0,
      "votes": 3
    },
    {
      "id": 369516,
      "title": "How is test-set extracted?",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 368747,
      "title": "💡How to ensemble predictions -- a key component to every strong solution 🏅",
      "comment_count": 2,
      "views": 0,
      "votes": 12
    },
    {
      "id": 368732,
      "title": "How do you train Ranking model?",
      "comment_count": 10,
      "views": 0,
      "votes": 14
    },
    {
      "id": 368541,
      "title": "What are we predicting?",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 364895,
      "title": "What to to learn about recommendation engine",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 368384,
      "title": "How to train a Word2Vec model 🚀🚀🚀",
      "comment_count": 4,
      "views": 0,
      "votes": 27
    },
    {
      "id": 367783,
      "title": "Ignore cheaters when using leaderboard",
      "comment_count": 1,
      "views": 0,
      "votes": 6
    },
    {
      "id": 364216,
      "title": "💡A robust local validation framework 🚀🚀🚀",
      "comment_count": 5,
      "views": 0,
      "votes": 31
    },
    {
      "id": 368385,
      "title": "💡How to improve the results of your Approximate Nearest Neighbor search! (annoy)",
      "comment_count": 0,
      "views": 0,
      "votes": 29
    },
    {
      "id": 368878,
      "title": "RMD file not loading my dataset",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 368149,
      "title": "Products in test and not in train?",
      "comment_count": 7,
      "views": 0,
      "votes": 17
    },
    {
      "id": 367754,
      "title": "How to thrive in this competition without going crazy (part 2 of 2) ❤️‍🔥",
      "comment_count": 3,
      "views": 0,
      "votes": 47
    },
    {
      "id": 366893,
      "title": "[Starter Pack] Matrix Factorization [Pytorch + Merlin Dataloader] 🚀🚀🚀",
      "comment_count": 6,
      "views": 0,
      "votes": 33
    },
    {
      "id": 366474,
      "title": "💡 Can you beat static rules with a ranker model without additional features?",
      "comment_count": 11,
      "views": 0,
      "votes": 24
    },
    {
      "id": 363624,
      "title": "Andrew Ng Recommender Systems",
      "comment_count": 15,
      "views": 0,
      "votes": 62
    },
    {
      "id": 363874,
      "title": "A top-down perspective on the current metric values",
      "comment_count": 24,
      "views": 0,
      "votes": 46
    },
    {
      "id": 368156,
      "title": "Unseen Aids in Test Set",
      "comment_count": 4,
      "views": 0,
      "votes": 5
    },
    {
      "id": 366477,
      "title": "📖 What are some good resources to learn about how gradient boosted tree ranking models work?",
      "comment_count": 6,
      "views": 0,
      "votes": 22
    },
    {
      "id": 367870,
      "title": "We can try more NLP methods in sequence recomendation, like ....",
      "comment_count": 0,
      "views": 0,
      "votes": 10
    },
    {
      "id": 364223,
      "title": "How to incorporate event-type with DL models/embeddings?",
      "comment_count": 1,
      "views": 0,
      "votes": 7
    },
    {
      "id": 367463,
      "title": "What happened so far? Full list of resources and discussions for a quick start to the competition 🚀",
      "comment_count": 5,
      "views": 0,
      "votes": 21
    },
    {
      "id": 365153,
      "title": "💡 Best hyperparams for the co-visitation matrix based on HPO study with 30 runs",
      "comment_count": 2,
      "views": 0,
      "votes": 24
    },
    {
      "id": 367282,
      "title": "A Case Study - How Youtube approaches the Candidate Generation and Ranking problems?",
      "comment_count": 1,
      "views": 0,
      "votes": 22
    },
    {
      "id": 368258,
      "title": "Some caution to apply NLP methods",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 367495,
      "title": "The Cold Start Problem",
      "comment_count": 0,
      "views": 0,
      "votes": 16
    },
    {
      "id": 367058,
      "title": "from zero to 60 in 2 seconds or less 🏎️🚓🚓🚓",
      "comment_count": 0,
      "views": 0,
      "votes": 22
    },
    {
      "id": 367588,
      "title": "Explainability for Multi-Objective Recommender Systems",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 367313,
      "title": "Segmentation Session or Product",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 366962,
      "title": "Does entering a submission mean you are required to share your code? ",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 365290,
      "title": "Sample Submission Questions",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 364062,
      "title": "📈 What do we know so far? ⚡Summary with  links to relevant resources",
      "comment_count": 7,
      "views": 0,
      "votes": 82
    },
    {
      "id": 366194,
      "title": "[Starter pack] LGBMRanker with polars 🚀🚀🚀",
      "comment_count": 2,
      "views": 0,
      "votes": 34
    },
    {
      "id": 366704,
      "title": "The World of RecSys",
      "comment_count": 0,
      "views": 0,
      "votes": 10
    },
    {
      "id": 363658,
      "title": "There are some bugs in evaluation",
      "comment_count": 17,
      "views": 0,
      "votes": 58
    },
    {
      "id": 366478,
      "title": "Q about the Data",
      "comment_count": 3,
      "views": 0,
      "votes": 7
    },
    {
      "id": 365716,
      "title": "TOP Resources |  Multi-Objective Recommender System",
      "comment_count": 1,
      "views": 0,
      "votes": 15
    },
    {
      "id": 366711,
      "title": "Add to cart without clicks, purchase without clicks/cart event are valid events?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 366169,
      "title": "How events in test sessions are split?",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 365917,
      "title": "RePlay - opensource RecSys constructor",
      "comment_count": 0,
      "views": 0,
      "votes": 18
    },
    {
      "id": 364375,
      "title": "💡 Do not disregard longer sessions -- they contribute disproportionately to the competition metric!",
      "comment_count": 1,
      "views": 0,
      "votes": 23
    },
    {
      "id": 363814,
      "title": "Resources for getting started with recommender systems",
      "comment_count": 16,
      "views": 0,
      "votes": 47
    },
    {
      "id": 365589,
      "title": "Co-Visitation Matrices and Matrix Factorization",
      "comment_count": 0,
      "views": 0,
      "votes": 25
    },
    {
      "id": 365907,
      "title": "JSONL files transformed to CSV format",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 365395,
      "title": "Can we summit prediction contains Duplicate aids?",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 365554,
      "title": "Test Set Clarification Questions",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 364210,
      "title": "Co-visitation matrix with simplified code and improved logic! 🚀 [LB 0.558]",
      "comment_count": 4,
      "views": 0,
      "votes": 31
    },
    {
      "id": 364722,
      "title": "🐘 the elephant in the room -- high cardinality of targets and what to do about this",
      "comment_count": 2,
      "views": 0,
      "votes": 20
    },
    {
      "id": 364530,
      "title": "Which metric is correct?",
      "comment_count": 5,
      "views": 0,
      "votes": 29
    },
    {
      "id": 365094,
      "title": "What is JSONL data type?",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 364534,
      "title": "📅 Dataset for local validation created using organizer's repository (parquet files)",
      "comment_count": 1,
      "views": 0,
      "votes": 34
    },
    {
      "id": 365039,
      "title": "time stamps on videos for getting started on recsys",
      "comment_count": 0,
      "views": 0,
      "votes": 8
    },
    {
      "id": 365125,
      "title": "Otto recommender Kernel Stats",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 364064,
      "title": "The order of appearance of type in each aid",
      "comment_count": 13,
      "views": 0,
      "votes": 32
    },
    {
      "id": 364942,
      "title": "Where is Customer and Products dataset for Customers Profiling and Products Profiling",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 364639,
      "title": "Papers and Winning Code From NVidia Recommender Systems Team",
      "comment_count": 0,
      "views": 0,
      "votes": 18
    },
    {
      "id": 363554,
      "title": "what does a session mean?",
      "comment_count": 10,
      "views": 0,
      "votes": 32
    },
    {
      "id": 364697,
      "title": "Order without clicking and adding to cart?",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 364176,
      "title": "Are there identical users in Train and Test data?",
      "comment_count": 6,
      "views": 0,
      "votes": 16
    },
    {
      "id": 364731,
      "title": "I have question about cart and order objective",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 363792,
      "title": "Ground-truth?",
      "comment_count": 6,
      "views": 0,
      "votes": 16
    },
    {
      "id": 364229,
      "title": "Why best model might not win RecSys competition",
      "comment_count": 1,
      "views": 0,
      "votes": 20
    },
    {
      "id": 363617,
      "title": "Processed dataset",
      "comment_count": 6,
      "views": 0,
      "votes": 30
    },
    {
      "id": 364228,
      "title": "Working as Vanga: which model would win this competition?",
      "comment_count": 1,
      "views": 0,
      "votes": 8
    },
    {
      "id": 363980,
      "title": "transformers4rec - Sequential and Session-Based Recommendation Systems",
      "comment_count": 4,
      "views": 0,
      "votes": 31
    },
    {
      "id": 364260,
      "title": "Some business specific questions about the products(aids) and website.",
      "comment_count": 0,
      "views": 0,
      "votes": 15
    },
    {
      "id": 363772,
      "title": "Metric Issue Resolved",
      "comment_count": 1,
      "views": 0,
      "votes": 38
    },
    {
      "id": 364377,
      "title": "[BUG] WRONG OTTO URL (MAYBE)",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 363638,
      "title": "Is it time to change the evaluation metric?",
      "comment_count": 11,
      "views": 0,
      "votes": 25
    },
    {
      "id": 364265,
      "title": "Recsys22/Booking: Sequence modelling",
      "comment_count": 0,
      "views": 0,
      "votes": 8
    },
    {
      "id": 364095,
      "title": "500+ public repositories for Recommendation system-[with Resources]",
      "comment_count": 1,
      "views": 0,
      "votes": 12
    },
    {
      "id": 363962,
      "title": "Recall@20 Explained",
      "comment_count": 0,
      "views": 0,
      "votes": 21
    },
    {
      "id": 364089,
      "title": "Fun bit of trivia",
      "comment_count": 1,
      "views": 0,
      "votes": 12
    },
    {
      "id": 363671,
      "title": "Checking the metric",
      "comment_count": 3,
      "views": 0,
      "votes": 30
    },
    {
      "id": 364022,
      "title": "Guidance on what algos to try from one of the authors of Microsoft Recommenders repo",
      "comment_count": 0,
      "views": 0,
      "votes": 15
    },
    {
      "id": 363730,
      "title": "recommenders - Best Practices on Recommendation Systems by Microsoft",
      "comment_count": 1,
      "views": 0,
      "votes": 19
    },
    {
      "id": 363950,
      "title": "Top Solutions of Last Recommender Systems Competition",
      "comment_count": 1,
      "views": 0,
      "votes": 17
    },
    {
      "id": 363540,
      "title": "🏆🏁💵Some Nice resources on Recommender systems for this contest🔮💰🏆",
      "comment_count": 3,
      "views": 0,
      "votes": 27
    },
    {
      "id": 363603,
      "title": "RecBole - A framework for recommendation",
      "comment_count": 1,
      "views": 0,
      "votes": 31
    },
    {
      "id": 364099,
      "title": "Great Resource about Session-based Recommender Systems",
      "comment_count": 0,
      "views": 0,
      "votes": 9
    },
    {
      "id": 363869,
      "title": "Question about the definition of the ground truth",
      "comment_count": 4,
      "views": 0,
      "votes": 12
    },
    {
      "id": 363722,
      "title": "Recommendation System Tutorials & Presentations ",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 363925,
      "title": "The relatively less data for each test session",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 363519,
      "title": "I'm so Chunked!  Better say: Trying to deal with Chunks",
      "comment_count": 4,
      "views": 0,
      "votes": 15
    },
    {
      "id": 363834,
      "title": "How is session defined?",
      "comment_count": 1,
      "views": 0,
      "votes": 7
    },
    {
      "id": 363544,
      "title": "💥🔮🏆Some good training / learning resources on Recommender Systems🏆🔮💥",
      "comment_count": 2,
      "views": 0,
      "votes": 9
    },
    {
      "id": 363860,
      "title": "💥💥 BEST CV-LB Correlation thread 💥💥",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 363845,
      "title": "co-occurrance baseline (ephemeral 1st place baseline)",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    }
  ],
  "errors": []
}