\n-\tThe training set: ML-MATT-CompetitionQT2021_train.csv contains 36,904 samples, each having 13 features and a label. Note that there may be erroneous samples and outliers.
\n-\tThe test set: ML-MATT-CompetitionQT2021_test.csv contains 9,158 samples following the same structure as the training set but not including the labels.
\n\nYou will have to upload the predictions for the test set and Kaggle will compare your predictions with the ground-truth labels in order to compute a score.", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Description", "content": "****Please do not include any notebook in this competition****\n\n**1.\tIntroduction**\n\nThe purpose of this homework is to solve a classification problem proposed as a competition in the Kaggle InClass platform, where each team of two members will try to get the maximum score. You can apply any of the concepts and techniques studied in class for exploratory data analysis, feature selection and classification.\n\n**2.\tProblem description**\n\nContext:\nTraditionally, the design of a cellular network focuses on the optimization of energy and resources that guarantees a smooth operation even during peak hours (i.e. periods with higher traffic load). However, this implies that cells are most of the time overprovisioned of radio resources. Next generation cellular networks ask for a dynamic management and configuration in order to adapt to the varying user demands in the most efficient way with regards to energy savings and utilization of frequency resources. If the network operator were capable of anticipating to those variations in the users’ traffic demands, a more efficient management of the scarce (and expensive) network resources would be possible.\nCurrent research in mobile networks looks upon Machine Learning (ML) techniques to help manage those resources. In this case, you will explore the possibilities of ML to detect abnormal behaviors in the utilization of the network that would motivate a change in the configuration of the base station.\n\nGoal:\nThe objective of the network optimization team is to analyze traces of past activity, which will be used to train an ML system capable of classifying samples of current activity as:\n•\t**0 (normal)**: current activity corresponds to normal behavior of any working day and. Therefore, no re-configuration or redistribution of resources is needed.\n•\t**1 (unusual)**: current activity slightly differs from the behavior usually observed for that time of the day (e.g. due to a strike, demonstration, sports event, etc.), which should trigger a reconfiguration of the base station.\n\n\nContent:\nThe dataset has been obtained from a real LTE deployment. During two weeks, different metrics were gathered from a set of 10 base stations, each having a different number of cells, every 15 minutes. The dataset is provided in the form of a csv file, where each row corresponds to a sample obtained from one particular cell at a certain time. Each data example contains the following features:\n\n•\t**Time** : hour of the day (in the format hh:mm) when the sample was generated.\n•\t**CellName1**: text string used to uniquely identify the cell that generated the current sample. CellName is in the form xαLTE, where x identifies the base station, and α the cell within that base station (see the example in the right figure).\n•\t**PRBUsageUL** and **PRBUsageDL**: level of resource utilization in that cell measured as the portion of Physical Radio Blocks (PRB) that were in use (%) in the previous 15 minutes. Uplink (UL) and downlink (DL) are measured separately.\n•\t**meanThr_DL** and **meanThr_UL**: average carried traffic (in Mbps) during the past 15 minutes. Uplink (UL) and downlink (DL) are measured separately.\n•\t**maxThr_DL** and **maxThr_UL**: maximum carried traffic (in Mbps) measured in the last 15 minutes. Uplink (UL) and downlink (DL) are measured separately.\n•\t**meanUE_DL** and **meanUE_UL**: average number of user equipment (UE) devices that were simultaneously active during the last 15 minutes. Uplink (UL) and downlink (DL) are measured separately.\n•\t**maxUE_DL** and **maxUE_UL**: maximum number of user equipment (UE) devices that were simultaneously active during the last 15 minutes. Uplink (UL) and downlink (DL) are measured separately.\n•\t**maxUE_UL+DL**: maximum number of user equipment (UE) devices that were active simultaneously in the last 15 minutes, regardless of UL and DL.\n\n•\t**Unusual**: labels for supervised learning. A value of 0 determines that the sample corresponds to normal operation, a value of 1 identifies unusual behavior.\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Evaluation", "content": "
The evaluation metric for this competition is Mean F1-Score. The F1 score, commonly used in information retrieval, measures accuracy using the statistics precision p and recall r. Precision is the ratio of true positives (TP) to all predicted positives (TP + FP). Recall is the ratio of true positives to all actual positives (TP + FN). The F1 score is given by:
\n\\[ F1 = 2\\frac{p \\cdot r}{p+r}\\ \\ \\mathrm{where}\\ \\ p = \\frac{TP}{TP+FP},\\ \\ r = \\frac{TP}{TP+FN} \\]
\nThe F1 metric weights recall and precision equally, and a good retrieval algorithm will maximize both precision and recall simultaneously. Thus, moderately good performance on both will be favored over extremely good performance on one and poor performance on the other.
\n