{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":8677062,"sourceType":"datasetVersion","datasetId":5201213}],"dockerImageVersionId":30732,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"**K-Nearest Neighbors** (KNN) is a versatile algorithm that extends *beyond traditional classification and regression* tasks to scenarios like fraud detection and hiring decisions. In such cases, KNN serves as a proximity-based method for **finding similar instances** in a dataset **without specific labels** or target variables. For instance, in **fraud detection**, KNN can identify transactions or behaviors resembling known fraudulent patterns by measuring similarity to historical fraud cases. Similarly, in **hiring decisions**, KNN can assist in identifying candidates with characteristics similar to successful past employees who held a particular position. This approach relies on the assumption that instances with similar features or attributes are likely to exhibit similar behaviors or outcomes, enabling KNN to provide valuable insights for decision-making in diverse problem domains. However, it's essential to consider the potential challenges, such as the need for *careful feature selection* and *scalability issues*, particularly with large datasets, when applying KNN in such contexts.","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19"}},{"cell_type":"markdown","source":"\nIn the given dataset containing **various features** influencing **car prices**, we encounter a scenario where new cars—a car and a truck—have been manufactured, and we aim to determine suitable prices for them based on the characteristics of the **most similar existing cars**. Leveraging the **K-Nearest Neighbors** (KNN) algorithm, we can identify the most similar cars in the dataset to the newly produced ones by computing their distances in the feature space. By selecting a suitable value for 'k,' the number of nearest neighbors to consider, we can then *average the prices of these similar cars to estimate the appropriate prices for the new car and truck*. ","metadata":{}},{"cell_type":"markdown","source":"### Read the datasets","metadata":{}},{"cell_type":"code","source":"import pandas as pd\n\n# Loading the dataset of priced cars from the specified CSV file into a DataFrame called PricedCars.\nPricedCars = pd.read_csv('/kaggle/input/cars-dataset/PricedCars.csv')\n\n# Displaying summary information about the PricedCars DataFrame, such as the column names, data types, and memory usage.\nPricedCars.info()\n\n# Loading the dataset of unpriced cars from the specified CSV file into a DataFrame called UnpricedCars.\nUnpricedCars = pd.read_csv('/kaggle/input/cars-dataset/UnpricedCars.csv')\n\n# Displaying summary information about the UnpricedCars DataFrame, such as the column names, data types, and memory usage.\nUnpricedCars.info()","metadata":{"trusted":true},"execution_count":26,"outputs":[{"name":"stdout","text":"\nRangeIndex: 2 entries, 0 to 1\nData columns (total 11 columns):\n # Column Non-Null Count Dtype \n--- ------ -------------- ----- \n 0 manufact 0 non-null float64\n 1 model 2 non-null object \n 2 price 0 non-null float64\n 3 engine_s 2 non-null float64\n 4 horsepow 2 non-null int64 \n 5 wheelbas 2 non-null float64\n 6 width 2 non-null float64\n 7 length 2 non-null float64\n 8 curb_wgt 2 non-null float64\n 9 fuel_cap 2 non-null float64\n 10 mpg 2 non-null int64 \ndtypes: float64(8), int64(2), object(1)\nmemory usage: 304.0+ bytes\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### Prepare the Data","metadata":{}},{"cell_type":"code","source":"from sklearn.preprocessing import StandardScaler\n\n# Creating an instance of the StandardScaler class to scale the features.\nz_score = StandardScaler()\n\n# Standardizing the features of the priced cars dataset using the fit_transform method,\nPricedCars_scaled = z_score.fit_transform(PricedCars.iloc[:, 3:11])\n\n# Standardizing the features of the unpriced cars dataset using the transform method,\nUnpricedCars_scaled = z_score.transform(UnpricedCars.iloc[:, 3:11])","metadata":{"execution":{"iopub.status.busy":"2024-06-21T19:07:10.239093Z","iopub.execute_input":"2024-06-21T19:07:10.239622Z","iopub.status.idle":"2024-06-21T19:07:10.256151Z","shell.execute_reply.started":"2024-06-21T19:07:10.239585Z","shell.execute_reply":"2024-06-21T19:07:10.254675Z"},"trusted":true},"execution_count":27,"outputs":[]},{"cell_type":"markdown","source":"### Find the Nearest Neighbors","metadata":{}},{"cell_type":"code","source":"from sklearn.neighbors import NearestNeighbors\n\n# Creating an instance of the NearestNeighbors class with the specified parameters:\nmodel = NearestNeighbors(n_neighbors=5, radius=1.0, metric='minkowski', p=2, n_jobs=-1)\n\n# Fitting the model to the standardized features of the priced cars dataset.\nmodel.fit(PricedCars_scaled)\n\n# Finding the k-nearest neighbors for the unpriced cars using the standardized features.\n# The kneighbors method returns the distances and indices of the neighbors.\ndistance, index = model.kneighbors(UnpricedCars_scaled)\n\n# Retrieving the nearest neighbors for the new car from the priced cars dataset using the indices.\nnewcar_neighbors = PricedCars.iloc[index[0]]\n\n# Retrieving the nearest neighbors for the new truck from the priced cars dataset using the indices.\nnewtruck_neighbors = PricedCars.iloc[index[1]]","metadata":{"trusted":true},"execution_count":34,"outputs":[{"execution_count":34,"output_type":"execute_result","data":{"text/plain":" manufact model price engine_s horsepow wheelbas width \\\n104 Nissan Quest 26.399 3.3 170.0 112.2 74.9 \n91 Mercury Villager 22.510 3.3 170.0 112.2 74.9 \n61 Honda Odyssey 26.000 3.5 210.0 118.1 75.6 \n13 Buick LeSabre 27.885 3.8 205.0 112.2 73.5 \n100 Mercedes-Benz M-Class 35.300 3.2 215.0 111.0 72.2 \n\n length curb_wgt fuel_cap mpg \n104 194.8 3.991 20.0 21.0 \n91 194.7 3.944 20.0 21.0 \n61 201.2 4.288 20.0 23.0 \n13 200.0 3.591 17.5 25.0 \n100 180.6 4.387 19.0 20.0 ","text/html":"
\n\n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n
manufactmodelpriceengine_shorsepowwheelbaswidthlengthcurb_wgtfuel_capmpg
104NissanQuest26.3993.3170.0112.274.9194.83.99120.021.0
91MercuryVillager22.5103.3170.0112.274.9194.73.94420.021.0
61HondaOdyssey26.0003.5210.0118.175.6201.24.28820.023.0
13BuickLeSabre27.8853.8205.0112.273.5200.03.59117.525.0
100Mercedes-BenzM-Class35.3003.2215.0111.072.2180.64.38719.020.0
\n
"},"metadata":{}}]}]}