Interview Questions

Frank A. - "1) create the experimental and control groups. 2) Then calculate the proportion (mean) of the true conversion rates for both groups using the convert column which counts True as 1 and False as 0. This is their conversion rates 3) calculate the statistic of the two groups by subtracting the proportion and standardizing. 4) get the p-value and compare with 0.05. 5) conclude the difference is statistically significant if the p-value is less than 0.05 otherwise no statistical difference"See full answer

Connor W. - "This video is a duplicate of the other video in this lesson, "Design A/B test for New Campaign""See full answer

Vishal S. - "Product Understanding - Ads are what you see from companies as stories, posts, reels. Post are from users (connections). We have to design an experience which produces maximum engagement while generating ad revenue. Clarifying Questions - Is it specific to posts/stories/reels ? Is there an existing post to ads ratio or do we have to start from scratch? Is it specific to a device/OS? Is it specific to a region/user demographic? Assumption - Existing posts to ads ratio"See full answer

Takashi M. - "My approach is try to narrow down by 4W1H. When - When this happened? Still continuously happen the problem? Where - Is this happen specific region or entire customers? What - What kind of problem? One time big error or any data show the curve? Who - Is it all platform or specific platform (Web, iOS, Android) How - How serious problem? How many customer affected? Next step is to narrow down the causes is from internal factors or external factors. For internal factors: Any recent up"See full answer

Abinash S. - "While running the testloop I am getting an error RuntimeError: runningmean should contain 28 elements not 38. I think it's the difference between the categorical features in train and test. `"See full answer

Dinar M. - "I used array to append. np.array does not have append and every np.vstack recreates object again. import numpy as np class Centroid: def init(self, location, vectors): self.location = location # (D,) self.vectors = vectors # (N_i, D) class KMeans: def init(self, n_features, k): self.nfeatures = nfeatures self.centroids = [ Centroid( location=np.random.randn(n_features), vectors=[] # I"See full answer

Robert H. - "Looking up stats, it says 1 billion swipes per day. I would infer that is not just matches. A swipe could be left or right."See full answer

Wizzy B. - "URL> DNS over UDP> IP address > TCP handshake > HTTP(SSL handshake) req, responses, websites architectures for any query DB, servers."See full answer