Saturday, August 4, 2018

Assignment on graph processing using GraphX in Apache Spark

  1. Generate a un-directional graph RDD from a given graph data.
  2. Compute listed vertex-based similarity measures for all the pairs of nodes in label data file. These similarity measures are computed between two nodes by utilizing neighborhood and/or node information of both nodes. Common neighbors Jaccard coefficient Adamic/Adar Preferential Attachment
  3. Bonus Question: Link Prediction Model Using the measures generated from the graph and labels from the labeled data to predict the possibility of new link formation. Please use the following steps. 1. Create a dataset by combining measures ( as features) and class labels from the labeled data. 2. Use decision tree based algorithm in SparkML to train the prediction model. 3. Split the dataset to generate the training data and testing data. 4. Use training data to build model and testing data to evaluate the model. 5. Present the model performance metrics: Accuracy, Recall, and Precision.
  1. Please follow the program submission instructions ( same as the previous assignments)
  2. Must use spark and GraphX for generating measures and use SparkML for bonus questions.
  3. More explanation on above graph measures here (
Graph data:
Use this link ( dl=0) to download graph data and file is formatted as shown below:

Label data:
click here (h ?dl=0) to download label data

How To Upgrade (Flash) Linksys' WRT54G/GL/GS Firmware to Tomato Firmware For IP Address and Bandwidth Monitoring

As a System Administrator one is usually faced with the challenge of providing the best possible technology solution within the confinem...