MonkeyLearn is a text analysis platform that helps you identify and extract actionable data from a variety of raw texts, including emails, chats, webpages, papers, tweets, and more! You can use custom tags to categorize texts, such as sentiments or topics, and extract specific data, such as organizations or keywords.
Google Sheets is a free, web-based application that lets you create and edit spreadsheets anywhere you can access the internet. Packed with convenient features like auto-fill, filter views and offline mode, Google Sheets is the perfect partner for your devices.Google Sheets Integrations
Google Sheets + Monkey LearnClassify Text in monkeylearn when New or Updated Spreadsheet Row is created in Google Sheets Read More...
Google Sheets + Monkey LearnExtract Text in monkeylearn when New or Updated Spreadsheet Row is created in Google Sheets Read More...
Google Sheets + Monkey LearnUpload training Data in monkeylearn when New or Updated Spreadsheet Row is created in Google Sheets Read More...
Google Sheets + Monkey LearnClassify Text in monkeylearn when New Spreadsheet is created in Google Sheets Read More...
Google Sheets + Monkey LearnExtract Text in monkeylearn when New Spreadsheet is created in Google Sheets Read More...
It's easy to connect Monkey Learn + Google Sheets without coding knowledge. Start creating your own business flow.
Triggers once a new spreadsheet is created.
Triggered when a new row is added to the bottom of a spreadsheet.
Trigger when a new row is added or modified in a spreadsheet.
Classifies texts with a given classifier.
Extracts information from texts with a given extractor.
Uploads data to a classifier.
Insert a new row in the specified spreadsheet.
Create a new spreadsheet row or Update an existing row.
Share Google Sheet.
Update a row in a specified spreadsheet.
Monkey Learn is a machine learning platform that allows users to create data models and then test their datasets with the models.
The platform offers three main types of Machine Learning models. Classifiers, Regressors, and Recommenders. As stated on the Monkeek Learn website, “These are all supervised-learning models, which means they require training data in order to learn how to generalize to new examples.”
Classifiers allow users to assign classes or categories based on features in a dataset. For example, using classifiers that are trained with positive and negative words, users can then use the datasets to create a bot that recognizes certain phrases as either positive or negative.
Regressors allow users to predict numeric values based on the features in the dataset. For example, using regressors trained with weather information on different days, users can then make an algorithm that predicts the day’s weather based on the features in the dataset.
Recommenders allow users to find similar examples on a dataset based on their features. For example, users can input a set of product features into the recommender model, and then the model will return similar products that have also had these features inputted into them. This can help users better understand what products are related to each other.
Google Sheets is a free top by Google that allows users to upload spreadsheets online for free editing. The top also offers “add-ons” which are additional tops for users to use with their spreadsheets. Add-ons can be integrated with other Google products such as Google Docs, Google Drive, Google Classroom, and more. Users can get add-ons from the Google Sheets store. Below is an image of an example of an add-on in action:
Google Sheets has an add-on feature that allows users to integrate other applications and services with their spreadsheet. Most notably, MonkeyLearn can be used in conjunction with Google Sheets through its add-on feature. MonkeyLearn provides two types of add-ons. one for when users want to train a model in the add-on (that will then be used in the add-on), and one when they want to use a model in the add-on (that was previously trained in MonkeyLearn. The fplowing image shows an example of the two types of add-ons:
In this example, we want to train a model in our add-on and use it in our Google Sheets spreadsheet. So we would choose the first option. “Use a Model from MonkeyLearn”. After choosing this option, we will be prompted to give the model a name. We will give it “My Model” and then click “Get Model”:
After clicking “Get Model”, we will be taken to the next step where we will be asked to select a template for how we want our model trained:
When creating a model in MonkeyLearn, users have two options for creating a model. a custom model or a default model. The default models are pre-made MonkeyLearn models that work well for specific tasks, whereas custom models are made from scratch by users and therefore need to be trained from scratch by the user before they can be used in MonkeyLearn or Google Sheets. The image below shows an example of a default model:
In this case, we will choose the “Classifier” template since we want our model to separate our dataset into two groups based on what it sees in our dataset. After selecting our template, we will click “Create Model”:
Here we will be able to name our model and set option parameters for how we want our model trained:
We will name our model “My Classifier” and then choose how many examples we want our model to have seen before being finished training it. 100% being all examples from the dataset. We will also choose how many examples we want our classifier to have seen before making predictions. 50% being half of all examples from the dataset. Then we will click “Continue”:
Next, we will see what type of data we want our classifier to use for training:
We will choose “Text” because we want our classifier to use text strings as its inputs. Next, we will choose what type of text data we want our classifier to use. “Free Text” or “Tags”. We will choose tags since we already have tags created for our dataset in MonkeyLearn. Then we will choose how many tags we want our classifier to use for training. 5 tags being 5 words from each example in our dataset. Next, if we want our classifier to use any additional data besides text like images or numbers (most likely no), then we can choose what additional data types we want it to use (we will leave this blank. We will then click “Continue”:
Now we can input what type of target variable (the one that is produced by the classifier. we want our classifier to predict (in this case it will be whether or not an email has spam):
We will make it predict whether or not an email has spam by choosing “yeso” as our target variable type (we could also have chosen “true/false” or another binary option. We will then choose how many times out of 10 our classifier needs to be correct on targets before being considered trained (we will choose 5. Then, if we want our classifier to stop learning after training it enough so that it performs well on validation data (we will leave this on), then we can check off either “Yes” or “No” (we will leave this on. Then we can choose if we want our classifier to learn faster by changing weights during training on some examples over others (we will leave this off. Finally, if there is any other data that might be useful for us to train our classifier on (like previous emails), then we can choose whether or not our classifier should learn from that data (we don’t have this data so we will leave this unchecked. We will then click “Continue”:
We now have enough information about our classifier so that it can start training itself on our dataset! We can see how quickly it is learning by watching the progress bar under the label “Training Progress”. If it takes longer than 30 minutes for us to see significant progress on the bar, then we might do one of two things. either increase the number of examples seen by at least 20%, or decrease how fast it learns so that it spends less time training itself but still works correctly. After seeing significant progress on the bar (a few hours to a day), we can now call up our classifier and test it out! We can view all of its options under the label “Available Options”:
We can now input text into one of these options and see what happens:
If you look closely at the screen capture above, you can see that I typed in this sentence. “I think this email might contain spam”. I put my cursor inside of the space between “I think” and “this email might contain spam” and clicked on my classifier option called “Label Spam” (which is circled in red above. This took me back to my spreadsheet where I saw that my classifier predicted that yes indeed my email did indeed contain spam:
The process to integrate 403 Forbidden and 403 Forbidden may seem complicated and intimidating. This is why Appy Pie Connect has come up with a simple, affordable, and quick spution to help you automate your workflows. Click on the button below to begin.