Ans i would like to select social networks ads

This preview shows page 5 - 6 out of 6 pages.

Ans: I would like to select social networks ads dataset for my semester topic. Scope: The scope of this analysis is to predict whether a person will buy a product displayed on a social network ad. 5 of 6 Location Dist_affected Damage_occure d No_ppl_died No_ppl_inure d Form_of_injur y Rescue_operation Rescue_officer No_ppl_resued Aid_provided
BIA 658 Social Network Analysis Vaibhavi Sahane Midterm Exam Spring 2021 2. Explain your dataset in terms of basic demographics (descriptive statistics with up to 5 attributes). What type of statistical analysis do you plan to perform and what software will you use? Ans: The analysis will be based on to find out product purchased the most by user’s attributes such as, UserID, Age, Gender, Estimated salary, and number of products purchased. The statistical analysis such as checking edges, checking the shortest path from a specific node, checking average short path length. I would like to use jupyter notebook for the statistical analysis and Gephi for the graphical representation of the data. 3. Using your dataset as a starting point, provide an example of unsupervised learning and a second example of supervised learning. You can either begin with unsupervised or supervised learning and then add more data to your example if needed. Ans: Supervised learning will work based on training data provided for supervised model. In this dataset, supervised model will predict the maximum products getting sold based on the data provided to it. For example, women over age of 40 tend to purchase the anti-aging cosmetics from the ads they watch on the social media, based on this data the supervised model will predict sales of anti-aging cosmetics based on previous records. This will help the company to recognize what kind of changes they should make to improve their number of targeted customers whether its improvement in marketing strategy, the product quality or the frequency of ads playing on the social media platform etc. in case of unsupervised learning, the prior training data would not be provided to the model, it will predict the possible sales outcome by learning the pattern by its own. Unsupervised learning follows a complex method as compared to supervised learning. 6 of 6

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture