{
"fetched_at": "2026-10-19T17:32:17Z",
"source_url": "https://www.kaggle.com/competitions/2024-artificial-intelligence-hw-1/data",
"pages": [
{
"name": "data-description",
"content": "## Files\n\n* **train.csv** - Contains approximately 9 years of daily stock price data for 1 stock. The labels to be predicted are 1_trend, 5_trend, and 10_trend, which represent the price trend (rise, decline, or unchanged) after the corresponding number of days.\n* **test.csv** - A dataset grouped by id, using 30 days of data to predict the trends of 1 stocks for 3 different future timeframes.\n* **sample_submission.csv** - A sample submission file showing the format. Please follow the format as outlined in the assignment instructions.\n\n## Columns\n* A brief description of the fields is as follows:\n\n| Indicator type | Column name |\n|:-----: | :-----: |\n| Momentum Indicators | ADX/ADXR/APO/AROON/AROONOSC/BOP/CCI/CMO/DX/
MACD/MACDEXT/MACDFIX/MFI/MINUS_DI/MINUS_DM/
MOM/PLUS_DI/PLUS_DM/PPO/ROC/ROCP/ROCR/ROCR100/
RSI/STOCH/STOCHF/STOCHRSI/TRIX/ULTOSC/WILLR |\n| Volume Indicators | AD/ADOSC/OBV |\n| Cycle Indicators | HT_DCPERIOD/HT_DCPHASE/HT_PHASOR/HT_SINE |\n| Price Transform | AVGPRICE/MEDPRICE/TYPPRICE/WCLPRICE |\n| Volatility Indicators | ATR/NATR/TRANGE |\n| Statistic Functions | BETA/ CORREL/ LINEARREG/ LINEARREG_ANGLE /
LINEAR_REG_INTERCEPT/LINEARREG_SLOPE/STDDEV/TSF/VAR |\n| Overlap Studies | BBANDS/ DEMA/ EMA/ HT_TRENDLINE/ KAMA/ MA/
MAMA/MAVP/MIDPOINT/MIDPRICE/SAR/
SAREXT/SMA/T3/TEMA/TRIMA/WMA |\n\n- Financial Environment Indicators: These include the opening, high, low, and closing prices of the Taiwan Stock Exchange index. These indicators can be used to assess whether money is flowing into or out of the stock market. The USD exchange rate can be used to check if money is flowing into Taiwan.\n- Momentum Indicators: Momentum indicators are used to assess the speed of price changes in order to determine the strength of stock prices and potential price reversals. These include indicators such as the Relative Strength Index (RSI) and the Moving Average Convergence Divergence (MACD), commonly used to capture changes in momentum within trends.\n- Volume Indicators: These indicators analyze changes in trading volume to assess the strength or weakness of a trend. Volume is considered a confirmation of price movement. Common indicators include Accumulation/Distribution Line (AD) and On-Balance Volume (OBV).\n- Cycle Indicators: Cycle indicators attempt to identify market cycles, which are periodic patterns of price increases and decreases. For example, the Hilbert Transform indicators (such as HT_DCPERIOD) are used to identify the length and phase of market cycles.\n- Price Transform: These indicators are used to create simplified or composite price models based on various price data points (opening price, high price, low price, closing price). They are typically used to smooth price data and extract trend information.\n- Volatility Indicators: Volatility indicators measure the magnitude of price fluctuations in the market, helping investors understand market uncertainty and risk levels. The Average True Range (ATR) is a common volatility indicator.\n- Statistic Functions: Statistical indicators use basic mathematical and statistical methods to analyze and simulate market behavior. This includes regression analysis, correlation analysis, etc., which are used to predict future market trends or assess relationships between variables.\n- Overlap Studies: These indicators are mainly used to analyze market trends and support/resistance levels. They often overlay price charts in the form of moving averages or other indicators to help identify price trends. Examples include the Simple Moving Average (SMA) and the Exponential Moving Average (EMA).\n\n* [Column Detail Link](https://hackmd.io/@aed58890/Sk2DIOL6C)",
"isPublished": false,
"hasIsPublished": false,
"postTitle": "",
"hasPostTitle": false,
"mimeType": "",
"hasMimeType": false
},
{
"name": "Description",
"content": "## Task\nUse 30 consecutive days stock data to predict the stock closing price trend after 1 day, 5 days, and 10 days. The trend can be one of three labels: rise, unchanged, or decline.\nYou need to make predictions using the provided dataset.\n\n## Train data\nThe data was scraped from FinMind, containing approximately 8 years and 1 stock of price data, along with 84 various financial indicators.\nEvery record contain 93 columns. See the columns detail in \"Data\" page.\n\n## Test Data\nTest data takes 30 consecutive days as a group. The order between groups has been shuffled, and the date is hidden.",
"isPublished": false,
"hasIsPublished": false,
"postTitle": "",
"hasPostTitle": false,
"mimeType": "",
"hasMimeType": false
},
{
"name": "Evaluation",
"content": "F1 Score\n\n[scikit-learn](https://scikit-learn.org/stable/modules/generated/sklearn.metrics.f1_score.html)\n### Submission format\n- Please refer to sample_submission.csv\n- Set team name as your student ID",
"isPublished": false,
"hasIsPublished": false,
"postTitle": "",
"hasPostTitle": false,
"mimeType": "",
"hasMimeType": false
},
{
"name": "rules",
"content": "Any form of cheating is not permitted, including but not limited to:\n\n- Copying code from classmates (including from students who have taken the course in the past)\n- Searching online or manually labeling data\n- Registering multiple accounts on Kaggle to upload files\n- Uploading data to previous Kaggle competitions",
"isPublished": false,
"hasIsPublished": false,
"postTitle": "",
"hasPostTitle": false,
"mimeType": "",
"hasMimeType": false
},
{
"name": "abstract",
"content": "Artificial Intelligence CE6020 HW1: Stock Trend Prediction",
"isPublished": false,
"hasIsPublished": false,
"postTitle": "",
"hasPostTitle": false,
"mimeType": "",
"hasMimeType": false
}
]
}