Embedded Files
Yosef Worku Alemneh
Highlights
Highlights
Amharic Speech To Text
Amharic Speech To Text
Built a deep learning model that is capable of transcribing a speech to text in the Amharic language. The model had Convolutional, DNN, and LSTM layers an used the CTC loss as its loss function.
APPROACH
Used 10,000 audio clips, and transcripts of the audios to train a deep learning model. The input to the model would be an audio sampled at 44.1 kHz, and the output would be a transcript.
The model had 4 convolutional layers, followed by two fully connected layers, two LSTM layers, a fully connected output layer
Used the Connectionist Temporal Classification (CTC) loss function and the Word Error Rate (WER) as a metric.
Pharmaceutical Sales Prediction
Pharmaceutical Sales Prediction
The finance team of a pharmaceutical chain that has 1115 stores wants to forecast sales in all their 1115 stores across several cities six weeks ahead of time. Used Linear Regression, Random Forest Regression, and LSTM Regression to build an end-to-end product that delivers this prediction to analysts in the finance team.
APPROACH
After performing univariate analysis and feature engineering, we trained three types of ML models for predicting sales.
Linear Regression - achieved an r2 score of 0.56
Random Forest Regression - achieved an r2 score of 0.91
LSTM Regression - trained one model for each store and achieved an r2 score of 0.94
Telecom Customer Data Analysis
Telecom Customer Data Analysis
Analyzed the customer information of a mobile service provider company. Provided a report to analyze opportunities for growth and made a recommendation on whether the company is worth buying or selling.
APPROACH
Performed Univariate and Mutivariate analysis
Ranked users based on their user engagement and user experience scores
Performed K-means clustering to classify users based on their satisfaction
A/B Testing
A/B Testing
A/B testing is a user experience research methodology. A/B tests consist of a randomized experiment with two variants, A and B., which are identical except for one variation that might affect a user's behavior. It includes application of statistical hypothesis testing or "two-sample hypothesis testing" as used in the field of statistic
A/B TESTING
A/B TESTING
Metric Choice:
Invariant metrics-Used this to ensure that the experiment (the way we presented a change to a part of the population )is not inherently wrong. eg number of users in both groups
Evaluation metrics-metrics we expect to change and are relevant to the goals we aim to achieve eg (brand awareness) Hypothesis testing for A/B testing
We use hypothesis testing to test the two hypotheses: Null Hypothesis :There is no difference in brand awareness between the exposed and control groups in the current case. Alternative Hypothesis: There is a difference in brand awareness between the exposed and control groups in the current case.
Machine Learning
Machine Learning
Carried out 3 types of classification analysis to predict whether a user responds yes to brand awareness, namely: Logistic Regression Decision Trees XGboost ,then compared the different classification models to assess the best performing one(s).
Page updated
Google Sites
Report abuse