In Kaggle, you just need to submit your prediction result in '.csv' format, but pay attention to other materials that need to be submitted on the canvas platform.
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Description", "content": "PLEASE JOIN THIS COMPETITION BEFORE 2023-05-18
\nGNN has performed very well in node classification and link prediction of heterogeneous and isomorphic graphs. In many recommendation scenarios, the trust relationship between users and the essential attributes of products are used, combined with users' purchase records, and using the representation generated by GNN can also achieve good results. However, in the scenarios given by the academic platform, we can find that collaborator recommendations, paper recommendations, and reviewer recommendations for journals and conferences are the main tasks.
\n\nFor most of the above problems, there are relatively good benchmarks and baselines. However, in the paper recommendation scenario, if it is for scholars conducting scientific research through their collaborators and the paper community where their papers are located, citation recommendation still lacks better datasets and models. A dataset we share here is used in a recommender system. When both the user and product have an associated network, we can extract group features for the user group and product category. Based on the relation between products and the connections between users, it provides users with better and more various recommendations, and this may help to solve the problem of less user behavior caused by the cold start of the recommender system.
\n\nHere is a link prediction problem in an academic network. We collected 6,611 authors and corresponding 79,937 papers from top journals in the field of GeoScience as well as citation information of their publications. The collected information is used to form an academic network, and there is a feasible way:
\n\nBuild a heterogeneous network, which contains two types of nodes, one type of nodes represents authors, and the other represents papers. In this network, each edge between an author node and a paper node means that the authors have read the paper (connecting authors and the papers cited by the papers written by the authors), each edge between two author nodes denotes the co-authorship, and each directed edge between two paper nodes represents the citation relation.
\n\nYou can propose some other way to form an academic network. Note that what we provide is the author and paper information so far, and it reveals the correlation between work of different authors. Consequently, suppose that you are designing an academic reading recommendation system, and you need to pick out papers related to the author's previous research. This problem can be modelled as a link prediction problem, and your task is to predict each author-paper pairs in the test set based on the information provided. If the paper is recommended to the author, mark it as 1, otherwise mark it as 0.
\n\nData used in this project is provided by Acemap group.
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Evaluation", "content": "The evaluation metric for this competition is F1-Score. It measures accuracy using the statistics Precision and Recall. Precision is the ratio of true positives (tp) to all predicted positive samples (tp + fp). Recall is the ratio of true positives to all positive samples (tp + fn). The F1-Score is given by:
\nF(f) = \\frac{Precision \\cdot Recall}{Precision + Recall}]
\n\nBefore the submission deadline , you can only view the F1-Score evaluated on 50% of the dataset on the leaderboard. After the deadline, we will reveal your model F1-Score evaluated on the entire dataset to rate your performance. However, you do not need to worry that your score would be affected by 1 or 2 percent of F1-Score loss; we will mark based on the baselines your model defeats.\n
\nYour submission file should contain two columns: Id and Probability. Each test author node pair is allocated a unique id, and you need to use this id as your prediction index. For example, the first row 3280,50982 means a test node pair with author id 3280 and paper id 50982, its index is 0. If you predict there exists an edge between 3280 and 50982 with prediction 1, the value of the corresponding row of your uploaded file is 0,1. \n
The file should contain a header and have the following format:
\nIndex,Predicted\n
0,1
1,0
Detailed description can be found in ‘data’ page.
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "rules", "content": "1. Please do not copy someone else's code. We will run plagiarism check after submission.\n\n2. Please do not download the dataset somewhere else to train your model. We have shuffled the dataset and will reproduce your experimental results using the dataset. A violation would be considered if there is a big gap between your reported results and our reproduced results.\n\n3. Please do not use the pre-trained model. Everything should be built from scratch.\n\n4. You need to zip and upload the following files before the deadline: (1) Your codes. (2) A readme.txt, which includes the scripts or instructions to run your codes. (3) A research paper, write in conference format. The final zip file should be uploaded via our course website: [https://oc.sjtu.edu.cn](https://oc.sjtu.edu.cn)", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "References", "content": "1. 【arXiv 2020】 Graph neural networks in recommender systems: a survey. [paper](https://arxiv.org/pdf/2011.02260)\n2. 【arXiv 2021】 Graph learning based recommender systems: A review. [paper](https://arxiv.org/pdf/2105.06339)\n3. 【KDD 2018】 Graph Convolutional Matrix Completion (GC-MC). [paper](https://arxiv.org/pdf/1706.02263)\n4. 【KDD 2018】 Graph convolutional neural networks for web-scale recommender systems (PinSage). [paper](https://dl.acm.org/doi/pdf/10.1145/3219819.3219890?casa_token=oYvwvyblQqEAAAAA:PTnSdOyciaB1PGAyivQWyuLi--BNDQxIPvIC-Rnoueq0bIRph7bE6GQvAk_QSDnh63fU-4M8FbL475o)\n5. 【RecSys 2018】 Spectral collaborative fltering (SpectralCF). [paper](https://dl.acm.org/doi/pdf/10.1145/3240323.3240343?casa_token=P6_dH2xN8VgAAAAA:OoHTVx-AU2cnh_NS1mFxr2ud-bELXxpqLMLcYs0OXgngv3LAHVJl2n96MN30xQbgKTmAhGV62b__XMc)\n6. 【SIGIR 2019】 Neural graph collaborative filtering (NGCF). [paper](https://dl.acm.org/doi/pdf/10.1145/3331184.3331267?casa_token=92VbA9XH7DQAAAAA:jVdvF559IxNw_E1q0w2gRfNqi6_cu_c1wtfvU_5E7t-bicy4KjwCUG2RK3nXMq-3EMP7UjN4Zy42uqc)\n7. 【SIGIR 2020】 Lightgcn: Simplifying and powering graph convolution network for recommendation (LightGCN). [paper](https://dl.acm.org/doi/pdf/10.1145/3397271.3401063?casa_token=ePMNswpnqxsAAAAA:YKqIuSnQbRVivX6RSbvbQTfIJJk4R44w-1Pom94_zFSBveXduYQ4ZCLnW85GbitVEDQPM00Rt0XzOA0)\n8. 【SIGIR 2019】 A neural influence diffusion model for social recommendation (DiffNet). [paper](https://dl.acm.org/doi/pdf/10.1145/3331184.3331214?casa_token=HOO3PYyWptUAAAAA:_eb0fGkcJhGVg9pMHOCob8cMl6fGLHpKD1VlQPnW-7nFML4H_vmIymFKtjE4twCa-FNSZBvS5w7KYpo)\n9. 【WWW 2019】 Graph neural networks for social recommendation (GraphRec). [paper](https://dl.acm.org/doi/pdf/10.1145/3308558.3313488?casa_token=sHyLa1l9Cy4AAAAA:9r2N8Mf37h4DDIm01J_FuYa9Id8F1Ejs_p2UcmYRdRtl9u2jUk4Dm9vL9S1XBBEsav9dxRAjZoykoXA)\n10. 【WWW 2019】 Dual graph attention networks for deep latent representation of multifaceted social effects in recommender systems (DANSER). [paper](https://dl.acm.org/doi/pdf/10.1145/3308558.3313442?casa_token=njPVZPVaAnYAAAAA:G4LTaT8AK_P1-Zgqvr2gJkLOKxn4_c9-c39I-AZYT0ttdElNQmqxWJnymBjVXu9VnHChe__-0pC5ewQ)\n11. 【RecSys 2019】 Deep social collaborative filtering (DSCF). [paper](https://dl.acm.org/doi/pdf/10.1145/3298689.3347011?casa_token=2pHA-jvXwtwAAAAA:MqlWqZ8kHdQ1UmmtGanOt8Mpf7_lURBi96wMrVpVhwMf0Ge8msVaWCRk6bXj_Iij6tDzvv27nilo1KY)\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false } ] }