SAS Visual Analytics

8月 282020
 

Remember back to your early school days, singing with all your classmates “If you’re happy and you know it clap your hands!” and then we’d all clap our hands. Being happy back then was so simple. Today, it’s hard to get away from all the negative headlines of 2020! It’s [...]

Analyzing happiness data in 2020 was published on SAS Voices by Melanie Carey

8月 182020
 

As the operations director of any organization’s contact center, you’ll want to ensure that your employees have the right tools in their hands to deliver superior customer service ultimately leading to happy customers and improved revenue for your company. In a previous series on How SAS Visual Analytics’ automated analysis takes customer care to the next level, we saw how automated analysis used machine learning and natural language generation (NLG) to determine which factors were most important in predicting if telecommunications customers would be likely to upgrade to a different mobile plan. We then used this information to create a list of customers our customer care workers could contact to promote our new products. But what about making on-the-fly decisions about what offer(s) best supports a given customer’s needs? To provide support for this type of analysis, SAS recently introduced the automated prediction feature within SAS Visual Analytics on SAS Viya.

What is automated prediction?

Automated prediction, in less than a minute, runs several analytic models (such as decision trees, gradient boosting, and logistic and linear regression) on a specific variable of your choice. Most of the remaining variables in your dataset are automatically analyzed as factors that might influence your specified variable. They are called underlying factors. SAS then chooses the one model (champion model) that most accurately predicts your target variable. The model prediction and the underlying factors are then displayed. You can adjust the values of the underlying factors to determine how the model prediction changes with each adjustment.

Let’s look at how automated prediction works.

Here we have the same customer table we were working within our previous blog posts. This table contains 121 columns (i.e. variables) containing usage and demographic information from a subset of customers who have contacted our customer care centers. One of these columns is a flag that indicates whether that customer upgraded their plan or not. We’ll use this as our target variable.

We’ll right-click on the Upgrade Flag variable and choose Predict on a new page in our report (Figure 1 below).

Figure 1: Selecting automated prediction.

With the automated prediction feature, SAS Visual Analytics created this easy-to-use form in less than a few minutes—by using advanced analytic models and machine learning in the background—that can help predict whether a customer is likely to upgrade (Figure 2 below). SAS analyzed all the other variables in the dataset to determine which ones were most likely to influence our Upgrade Flag. We can modify some of the values in the form and see how it affects the outcome. The factors are listed in order of their relative importance.

Figure 2: Automated Prediction results.

Under the prediction, SAS provides further details using natural language generation (NLG) (Figure 3 below). For example, here we see that the most prevalent value in our data was Did not upgrade with 87.87% of the records having that value. We can also see what type of model was chosen as the champion model; the model that best predicts whether a customer will upgrade. In our current example, the Gradient Boosting model provides the most accurate results.

Figure 3: Natural Language Generation (NLG) provides details about the prediction.

Now, let’s say that a customer calls in. We can fill in the values for each of these parameters listed that are specific to that customer. The analysis is automatically updated using the model previously generated and provides an updated prediction.

Here we see that this customer is likely to upgrade (Figure 4 below). Now we can discuss other mobile plans with this customer.

Figure 4: Updated prediction based on values entered for a specific customer.

What if we want to learn more about the model behind the prediction and what variables were deemed the most important? We can find that information when we maximize the object (Figure 5 below).

Figure 5: Maximize to see model details.

Now we can see at the bottom the steps that were taken to create the champion model (step 2) (Figure 6 below). SAS ran a decision tree, logistic regression, and a gradient boosting model and found that the gradient boosting model provided the greatest accuracy (91.67% accurate in predicting upgrade flag).

Figure 6: Prediction model description.

If we select the Relative Importance tab, we can see the relative importance of each of the underlying factors (Figure 7 below).

Figure 7: Relative importance of underlying factors.

Total Days Over Plan has the greatest influence on our Upgrade Flag variable. Days Suspended Last 6M is the next most important variable whose impact is 43.97% of Total Days Over Plan (Figure 7 above).

Automated Prediction is a fast and easy way to gain an understanding of how variables influence a target and to consider “what if,” by seeing how modifying those underlying factors affects an outcome. Multiple models were run on the data and a champion model was chosen for us using machine learning. We were able to see the accuracy of the champion model and the relative importance of each influencer. All this within a few minutes! This provides a great foundation for moving forward to more advanced modeling techniques. For those who wish to have more control over the models, SAS Visual Analytics also provides capabilities to build and modify other advanced analytical models such as gradient boosting, linear and logistic regression, and decision trees. Furthermore, as your company’s analytic maturing increases additional products can be easily added on to provide even more model choices (Forest, Neural network, Support vector machine, etc.) and capabilities. SAS’ platform and products support the whole analytical life cycle from data preparation all the way through model deployment, model performance management, and decision intelligence.

Take the next step in learning more about SAS Visual Analytics on SAS Viya by signing up for a free trial!

How SAS Visual Analytics' automated prediction takes customer care to the next level was published on SAS Users.

8月 182020
 

As the operations director of any organization’s contact center, you’ll want to ensure that your employees have the right tools in their hands to deliver superior customer service ultimately leading to happy customers and improved revenue for your company. In a previous series on How SAS Visual Analytics’ automated analysis takes customer care to the next level, we saw how automated analysis used machine learning and natural language generation (NLG) to determine which factors were most important in predicting if telecommunications customers would be likely to upgrade to a different mobile plan. We then used this information to create a list of customers our customer care workers could contact to promote our new products. But what about making on-the-fly decisions about what offer(s) best supports a given customer’s needs? To provide support for this type of analysis, SAS recently introduced the automated prediction feature within SAS Visual Analytics on SAS Viya.

What is automated prediction?

Automated prediction, in less than a minute, runs several analytic models (such as decision trees, gradient boosting, and logistic and linear regression) on a specific variable of your choice. Most of the remaining variables in your dataset are automatically analyzed as factors that might influence your specified variable. They are called underlying factors. SAS then chooses the one model (champion model) that most accurately predicts your target variable. The model prediction and the underlying factors are then displayed. You can adjust the values of the underlying factors to determine how the model prediction changes with each adjustment.

Let’s look at how automated prediction works.

Here we have the same customer table we were working within our previous blog posts. This table contains 121 columns (i.e. variables) containing usage and demographic information from a subset of customers who have contacted our customer care centers. One of these columns is a flag that indicates whether that customer upgraded their plan or not. We’ll use this as our target variable.

We’ll right-click on the Upgrade Flag variable and choose Predict on a new page in our report (Figure 1 below).

Figure 1: Selecting automated prediction.

With the automated prediction feature, SAS Visual Analytics created this easy-to-use form in less than a few minutes—by using advanced analytic models and machine learning in the background—that can help predict whether a customer is likely to upgrade (Figure 2 below). SAS analyzed all the other variables in the dataset to determine which ones were most likely to influence our Upgrade Flag. We can modify some of the values in the form and see how it affects the outcome. The factors are listed in order of their relative importance.

Figure 2: Automated Prediction results.

Under the prediction, SAS provides further details using natural language generation (NLG) (Figure 3 below). For example, here we see that the most prevalent value in our data was Did not upgrade with 87.87% of the records having that value. We can also see what type of model was chosen as the champion model; the model that best predicts whether a customer will upgrade. In our current example, the Gradient Boosting model provides the most accurate results.

Figure 3: Natural Language Generation (NLG) provides details about the prediction.

Now, let’s say that a customer calls in. We can fill in the values for each of these parameters listed that are specific to that customer. The analysis is automatically updated using the model previously generated and provides an updated prediction.

Here we see that this customer is likely to upgrade (Figure 4 below). Now we can discuss other mobile plans with this customer.

Figure 4: Updated prediction based on values entered for a specific customer.

What if we want to learn more about the model behind the prediction and what variables were deemed the most important? We can find that information when we maximize the object (Figure 5 below).

Figure 5: Maximize to see model details.

Now we can see at the bottom the steps that were taken to create the champion model (step 2) (Figure 6 below). SAS ran a decision tree, logistic regression, and a gradient boosting model and found that the gradient boosting model provided the greatest accuracy (91.67% accurate in predicting upgrade flag).

Figure 6: Prediction model description.

If we select the Relative Importance tab, we can see the relative importance of each of the underlying factors (Figure 7 below).

Figure 7: Relative importance of underlying factors.

Total Days Over Plan has the greatest influence on our Upgrade Flag variable. Days Suspended Last 6M is the next most important variable whose impact is 43.97% of Total Days Over Plan (Figure 7 above).

Automated Prediction is a fast and easy way to gain an understanding of how variables influence a target and to consider “what if,” by seeing how modifying those underlying factors affects an outcome. Multiple models were run on the data and a champion model was chosen for us using machine learning. We were able to see the accuracy of the champion model and the relative importance of each influencer. All this within a few minutes! This provides a great foundation for moving forward to more advanced modeling techniques. For those who wish to have more control over the models, SAS Visual Analytics also provides capabilities to build and modify other advanced analytical models such as gradient boosting, linear and logistic regression, and decision trees. Furthermore, as your company’s analytic maturing increases additional products can be easily added on to provide even more model choices (Forest, Neural network, Support vector machine, etc.) and capabilities. SAS’ platform and products support the whole analytical life cycle from data preparation all the way through model deployment, model performance management, and decision intelligence.

Take the next step in learning more about SAS Visual Analytics on SAS Viya by signing up for a free trial!

How SAS Visual Analytics' automated prediction takes customer care to the next level was published on SAS Users.

8月 182020
 

As the operations director of any organization’s contact center, you’ll want to ensure that your employees have the right tools in their hands to deliver superior customer service ultimately leading to happy customers and improved revenue for your company. In a previous series on How SAS Visual Analytics’ automated analysis takes customer care to the next level, we saw how automated analysis used machine learning and natural language generation (NLG) to determine which factors were most important in predicting if telecommunications customers would be likely to upgrade to a different mobile plan. We then used this information to create a list of customers our customer care workers could contact to promote our new products. But what about making on-the-fly decisions about what offer(s) best supports a given customer’s needs? To provide support for this type of analysis, SAS recently introduced the automated prediction feature within SAS Visual Analytics on SAS Viya.

What is automated prediction?

Automated prediction, in less than a minute, runs several analytic models (such as decision trees, gradient boosting, and logistic and linear regression) on a specific variable of your choice. Most of the remaining variables in your dataset are automatically analyzed as factors that might influence your specified variable. They are called underlying factors. SAS then chooses the one model (champion model) that most accurately predicts your target variable. The model prediction and the underlying factors are then displayed. You can adjust the values of the underlying factors to determine how the model prediction changes with each adjustment.

Let’s look at how automated prediction works.

Here we have the same customer table we were working within our previous blog posts. This table contains 121 columns (i.e. variables) containing usage and demographic information from a subset of customers who have contacted our customer care centers. One of these columns is a flag that indicates whether that customer upgraded their plan or not. We’ll use this as our target variable.

We’ll right-click on the Upgrade Flag variable and choose Predict on a new page in our report (Figure 1 below).

Figure 1: Selecting automated prediction.

With the automated prediction feature, SAS Visual Analytics created this easy-to-use form in less than a few minutes—by using advanced analytic models and machine learning in the background—that can help predict whether a customer is likely to upgrade (Figure 2 below). SAS analyzed all the other variables in the dataset to determine which ones were most likely to influence our Upgrade Flag. We can modify some of the values in the form and see how it affects the outcome. The factors are listed in order of their relative importance.

Figure 2: Automated Prediction results.

Under the prediction, SAS provides further details using natural language generation (NLG) (Figure 3 below). For example, here we see that the most prevalent value in our data was Did not upgrade with 87.87% of the records having that value. We can also see what type of model was chosen as the champion model; the model that best predicts whether a customer will upgrade. In our current example, the Gradient Boosting model provides the most accurate results.

Figure 3: Natural Language Generation (NLG) provides details about the prediction.

Now, let’s say that a customer calls in. We can fill in the values for each of these parameters listed that are specific to that customer. The analysis is automatically updated using the model previously generated and provides an updated prediction.

Here we see that this customer is likely to upgrade (Figure 4 below). Now we can discuss other mobile plans with this customer.

Figure 4: Updated prediction based on values entered for a specific customer.

What if we want to learn more about the model behind the prediction and what variables were deemed the most important? We can find that information when we maximize the object (Figure 5 below).

Figure 5: Maximize to see model details.

Now we can see at the bottom the steps that were taken to create the champion model (step 2) (Figure 6 below). SAS ran a decision tree, logistic regression, and a gradient boosting model and found that the gradient boosting model provided the greatest accuracy (91.67% accurate in predicting upgrade flag).

Figure 6: Prediction model description.

If we select the Relative Importance tab, we can see the relative importance of each of the underlying factors (Figure 7 below).

Figure 7: Relative importance of underlying factors.

Total Days Over Plan has the greatest influence on our Upgrade Flag variable. Days Suspended Last 6M is the next most important variable whose impact is 43.97% of Total Days Over Plan (Figure 7 above).

Automated Prediction is a fast and easy way to gain an understanding of how variables influence a target and to consider “what if,” by seeing how modifying those underlying factors affects an outcome. Multiple models were run on the data and a champion model was chosen for us using machine learning. We were able to see the accuracy of the champion model and the relative importance of each influencer. All this within a few minutes! This provides a great foundation for moving forward to more advanced modeling techniques. For those who wish to have more control over the models, SAS Visual Analytics also provides capabilities to build and modify other advanced analytical models such as gradient boosting, linear and logistic regression, and decision trees. Furthermore, as your company’s analytic maturing increases additional products can be easily added on to provide even more model choices (Forest, Neural network, Support vector machine, etc.) and capabilities. SAS’ platform and products support the whole analytical life cycle from data preparation all the way through model deployment, model performance management, and decision intelligence.

Take the next step in learning more about SAS Visual Analytics on SAS Viya by signing up for a free trial!

How SAS Visual Analytics' automated prediction takes customer care to the next level was published on SAS Users.

7月 312020
 

In this SAS demo, I demonstrate how to prompt for a date range in a SAS Visual Analytics report. We walk through how to configure prompts using four types of control objects:

Slider (range)

Slider (single value)

Drop-down list

Text input

Then we cover how to define parameters and configure filters for report objects using the date prompted boundaries. These techniques are demonstrated using SAS Viya 3.4 SAS Visual Analytics 8.4 but can be applied to earlier releases.

I've also written accompanying articles in a four-part series. Below the video see the links to the articles and their corresponding timestamps in the video.

01:33 – 1st Example: Slider (Range) - How to prompt for a date range in a SAS VA report – Example 1 Slider (Range)

03:01 – 2nd Example: Slider (Single Value) - How to prompt for a date range in a SAS VA report – Example 2 Slider (Single Value)

14:04 – 3rd Example: Drop-down list - How to prompt for a date range in a SAS VA report – Example 3 Drop-down list

19:33 – 4th Example: Text input - How to prompt for a date range in a SAS VA report – Example 4 Text input

Other References

How to prompt for a date range in a SAS Visual Analytics Report - Four Part Series was published on SAS Users.

7月 142020
 

SAS, in partnership with Intel, is sponsoring the 2020 Esri User Conference taking place virtually on Jul. 13-16. The Esri User Conference is the world’s largest virtual Geographic Information System (GIS) event and will include three days of inspiring plenary and technical sessions of how GIS is making a difference [...]

Being an Esri User: Data visualization and location analytics was published on SAS Voices by Robby Powell

7月 092020
 

How have healthcare providers and governmental agencies predicted the fast-changing, potentially exponential increase in the need for medical services and equipment through the various stages of the COVID-19 pandemic? Mathematical techniques that attempt to model and understand the likely spread of the disease have been instrumental. The SEIR model is [...]

Are your hospital resources at risk of hitting capacity? was published on SAS Voices by Melanie Carey

6月 192020
 

Adverse outcomes, and the rapid spread of COVID-19, have accelerated research on all aspects of the disease. You may have found it overwhelming, and very time-consuming, to find relevant and specialized insights in all the scientific literature out there. To aid researchers in quickly identifying relevant literature about key topics [...]

Speed up your COVID-19 research with text analysis: step-by-step was published on SAS Voices by Melanie Carey

4月 102020
 

First introduced in SAS Visual Analytics 8.3, common filters are filters that can be shared between objects in your reports.

Common filter benefits include:

  • Easy to assign the same filter conditions to other report objects.
  • When you edit a common filter, it is updated everywhere that the common filter is used.
  • A common filter is available for the entire report, across pages.

Common filter limitations include:

  • Objects must share the same data source as the common filter definition.
  • Common filters are not available across multiple reports.

While the benefits speak for themselves, common filters also expedite designing reports and exploring data by quickly reusing the same filter conditions with only a few mouse clicks.

Let’s look at some examples of using common filters. In the screenshot below, I defined a filter to return data for the last 30 days. I converted that filter to a common filter and now the Last 30 Days common filter can be applied to any object in this report that uses the same data source. Then I applied it to the bottom treemap object.



In my second example, I parameterized a text input control to capture a user-defined value. Then I defined a parameter-driven filter for the bar chart object. Next, I converted that filter to a common filter named Text Input Contains Filter and applied it to the list table object below.



In my third example, I parameterized two drop-down list controls to capture date boundaries. Then I defined a parameter-driven filter for the waterfall chart object. Next, I converted that filter to a common filter named FromToWeekFilter and applied it to the key value object.



Pro tip: Embrace meaningful filter names. Recall that these common filters are associated with the data source for which they are defined. Being able to quickly identify a common filter definition by its name will save you additional mouse clicks in the long run.

Example 1: Last 30 Days Filter

Let’s pick up where we need to define the filters. Therefore, both objects have been added to the report page with roles assigned.

Next, select the line chart object and open the Filters pane. Click + New filter then select Advanced filter.

Now we will use the built-in filter conditions offered in SAS Visual Analytics. First, select the date/datetime data item, Day. Second, scroll down in the available conditions until you see Last 30 days, then double click on the condition to add it to the expression editor. Third, give the filter condition a meaningful name and as a final step, be sure the number of returned observations is expected. Click OK.

Now we need to convert this filter to a common filter. With the line chart object still selected, on the Filters pane, use the Last 30 Day overflow menu and select Change to common filter.



You should now see the Last 30 Days listed as a Common Filter in the Data pane.

Lastly, we can apply this common filter to the treemap object. Select the treemap object, then open the Filters pane. Next click + New Filter and select the common filter Last 30 Days.



It’s good practice to title your objects to reflect any filters that may be defined for them. Especially if there are no prompts, i.e. control objects, driving the subset of data. This is so your report consumers quickly understand that they are only seeing, in this case, the last 30 days of data.

Example 2: Text Input Contains Filter

This next example will define a parameter-driven common filter. If you are not familiar with using parameters in SAS Visual Analytics, start with this article and refer to the additional materials at the end.

See the screenshot below for more information on the data item role assignments for each object. The most important role that will drive the parameter-driven common filter is the parameter. The parameter, TextInputParameter, is a character parameter assigned to the text input control object and it is the only role assignment. If there is nothing entered in this prompt, i.e. control object, then the filter will return all of the data. We will define a contains expression to only return data rows that contain the entered text.

With the bar chart object selected, open the Filters pane and click + New filter and select Advanced filter.



Now we need to build our parameter-driven filter expression. In this filter we will be checking if the Product Description data item Contains the entered text stored in the TextInputParameter. I wrapped each string expression in an UpCase function so that mixed case is ignored. I could have just as easily used the LowerCase function to get the same result.

And finally, remember to give your filter a meaningful name. The returned observations number is not reflective of an applied filter since I do not have any text entered in the control object. If you had text entered there, then your returned observation number may show a subset of matched rows.



Now we need to convert this object-level filter to a common filter. With the bar chart still the active object, open the Filters pane and use the Text Input Contains Filter overflow menu and select Change to common filter.

You should now see the Text Input Contains Filter in the Data pane.



To add this common filter to the lower list table object, select the list table object and open the Filters pane. Click on the + New filter and select the Text Input Contains Filter.



Now let’s test the filter. In the screenshot below, I’ve typed the word red. Recall that our filter expression is to return rows where Product Description contains the entered text. Notice that even though I do not have the data item Product Description assigned as a data role in the bar chart object, SAS Visual Analytics is still able to apply the filter appropriately. The list table object does have a role assigned for Product Description so that filter application is easier to identify.



I used the contains operator instead of the equals to return a partial match to the Product Description to help identify trends across multiple Products. In these next examples, I entered the text (F) and (M) to return the rows where the Product Description indicates gender-specific products.

Example 3: From – To Week Filter

This last example will also define a parameter-driven common filter. If you need a more step-by-step guide to similar examples, refer to this YouTube video tutorial:

See the screenshot below for more information on the data item role assignments for each object. This example differs from the last in that these control objects, the drop-down lists, use both Category and Parameter Role assignments. The Year-Week data item will provide the available values and the selections will be stored in the parameters FromWeekParameter and ToWeekParameter. I will then create a common filter for an inclusive between of the selected year week values to apply to both the key value object and the waterfall chart.



With the waterfall chart object selected, open the Filters pane and click + New filter and select Advanced filter.



Now we need to build our parameter-driven filter expression. In this filter, I will subset the Year-Week date values which are inclusively between the FromWeekParameter and ToWeekParameter boundaries. Follow these steps:

  1. Select the date/datetime data item, Year-Week.
  2. Scroll down in the available conditions till you see Year-Week BetweenInclusive(‘x’,’y’), then double click on the condition to add it to the expression editor.
  3. Drag the FromWeekParameter and ToWeekParameter from the parameter data items list to the expression.
  4. Give the filter condition a meaningful name. Click OK.



Test the filter by adjusting the values in the drop-down list controls.



Now we need to convert this object-level filter to a common filter. With the waterfall chart still the active object, open the Filters pane and use the FromToWeekFilter overflow menu and select Change to common filter.

You should now see the FromToWeekFilter in the Data pane.



To add this common filter to the key value object; select the key value object and open the Filters pane. Click on the + New filter and select the FromToWeekFilter.

Conclusion

Remember that once a common filter is defined, it can be used for any object in the report that uses the same data source. In these examples, I applied the common filter on the same report page but you can use a common filter on any page within the report. Recall that all of the common filters are listed and available when I click to apply a new filter to an object.

Hopefully these examples have shown you new ways to explore data faster!

Additional materials for using parameters in SAS Visual Analytics:

Using common filters in SAS Visual Analytics was published on SAS Users.

3月 112020
 

In the early days and weeks of any widespread global health concern, particularly in a fast-moving outbreak like the coronavirus, there are many unknowns. Data visualization can be a good starting point to understand trends and piece data points together into a meaningful story. The ability to visualize the spread [...]

Using data visualization to track the coronavirus outbreak was published on SAS Voices by Mark Lambrecht