Use Cases for
See Use Cases Below
01. NoCodeMLApp Text Analytics for Assessing Customer Sentiment in Marketing
You can use the NoCodeMLApp for Sentiment Analysis or Emotion Analysis or any type of text analytics that requires supervised learning. There are two MLTextAnalysis options in the NoCodeMLApp. One uses training data stored in a .csv or .json file.The other uses Directory-based approach. In this section, we look at the .csv or .json file-based option. The Directory-based option is discussed below in Section 02. Please click on the video on the left-side to see a demo of the Machine Learning Text Analysis (MLTextAnalysis) for classifying sentences into positive or negative sentiments.
02. NoCodeMLApp Directory-Based Text Analytics for Customer Engagement Analysis
In this use case, you will see how to use MLTextAnalysis for Sentiment Analysis but instead of using trained data file in .csv or .json format you will use Directory-based approach. What is Directory-based approach? In the Directory-based approach you label each directory with the name of the category. For example, you create Directory (or Folder) called "positive_sentiment" and another Directory called "negative_sentiment". Then in the directory called "positive_sentiment" you place sample text files containing "positive" sentiment text in .csv or .json format. You do NOT need to label individual text files. Similarly you place sample text files containing "negative" sentiment text in .csv or .json format. Again you do NOT have to label individual files. Once you point the upper level Directory containing both the "positive_sentiment" and "negative_sentiment" sub-directories to the NoCodeMLApp MLTextAnalysis - Directory-based option, the algorithm will train itself based the training data. For a full demonstration, please check out the video on the left side.
03. Image Classification Directory-Based Approach to Classify Objects for Use in Augmented Reality Mobile App Dev
You can perform image analysis and train your image classifier without writing a single line of code. All you need is to place images in Directories or Folders. For example, if you want to classify various types of fruits and train your Machine Learning Model all you have to do is place unlabeled images of various fruits say banana, dates, apricot, etc. in named folders. You place bunch of unlabeled banana images in 'Banana' folder then bunch of dates images in 'Dates' folder and a bunch of apricot images in 'Apricot' folder.
You place all of these named folders under a common folder called 'Fruits'. Then you choose the MLImageClassification option and point to the top level "Fruits" folder which now contains the subfolders for 'Banana', 'Dates' and "Apricot". The MLImageClassifier will learn to classify images based on Folder names. Instead of reading about it, just click on the video to the left for a demo.
04. Training: Using NoCodeMLApp To Learn Fundamentals of Machine Learning Workflow
NoCodeMachineLearningApp is simple to use. One can use this App to learn the fundamental workflow related to Machine Learning. One can focus on the Problem Domain and select Data that holds potential answers to important questions. One can clean the Data and then use the NoCodeMLApp to train a Machine Learning Model and evaluate its performance by looking at the Tables of useful Performance Metrics. It is easy to iterate through multiple training and evaluation sessions and save both performance metrics and test results for later analysis.
05. Potential Application of the NoCodeMLApp in Education
It is quite clear today that we need more educated and trained Professionals in every discipline to analyze available massive Data Sets to make more informed and effective decisions. Healthcare Data has hidden patterns that can help in early diagnosis, patient outcome, better treatment, reduce cost of Healthcare providers, etc. More professionals are neded who can be trained in using easy to use Machine Learning tools to perform initial analysis of these Data Sets. Typically Healthcare Professionals do have the time or opportunity to learn the underlying programming languages. But they certainly have massive expertise in Hel