social media data
Description:
Assessment 2 is an individual assessment which tests students’ ability to analyse social media data using NLP techniques and statistical methods.
The deliverables for this assessment: 1. Presentation on week 10 on part A of the project. (20% presentation) 2. Final project code and report for both parts A and Bon week 13. (60%)
The presentation will help stu
d presentation skills. sentation feedback is given by tutor and peers. rough this assessment, the student is required to: 1. Extract social media data e.g., Twitter, Facebook, YouTube. 2. Clean the collected data. 3. Apply appropriate statistical techniques for topic modelling/NLP to detect a group of words that best represent the information in the collection. 4. Process data to reveal new and interesting insights into the data, which may include recurring patterns of words in the text that may translate to the interestingness of the patterns. 5. Detect sentiments in the text that may determine the trends and topics. 6. Present your findings in a presentation and a technical report.
nvey analytics insights to the general
em in articulating their ancing their communication
The assessment consists of 2 parts; pa five data analyti • while part B focuses on text data analytics. Details are as Part A: Statistical analysis This part will focus on the statistical analysis of trends on social media. Students will use APIs to collect data from Twitter and Facebook to answer the following questions: 1. What are popular trends on Twitter at the moment, either in the UK or worldwide? Extract some insights from these trends such as: when it started in each place? What devices are used to tweet? and what sources can you trust? Use plots, graphs and maps to explain your insights. 2. Use one of the graph datasets available in Stanford Large Network Dataset Collection (https://snap.stanford.edu/data/) or any other publicly available graph dataset (you can also create your own graph), apply the following: a. Find the most important nodes (individuals) in the network based on different centrality measures, b. Visualise your graph using one of centrality measures of your choice, and c. Apply a Community Detection Algorithm to the graph, visualise the communities and discuss your findings.
2:30
CM P7202 Coursework Assess…
AssignmentTutorOnline