Fix typos in the Customer Segmentation case study (#957)

Signed-off-by: Kuan-Hao Huang <[email protected]>

Author: KuanHaoHuang

notebooks/CustomerScenarios/Case Study - Customer Segmentation at An Online Media Company - EconML + DoWhy.ipynb

@@ -45,7 +45,7 @@
     "\n",
     "The global online media market is growing fast over the years. Media companies are always interested in attracting more users into the market and encouraging them to buy more songs or become members. In this example, we'll consider a scenario where one experiment a media company is running is to give small discount (10%, 20% or 0) to their current users based on their income level in order to boost the likelihood of their purchase. The goal is to understand the **heterogeneous price elasticity of demand** for people with different income level, learning which users would respond most strongly to a small discount. Furthermore, their end goal is to make sure that despite decreasing the price for some consumers, the demand is raised enough to boost the overall revenue.\n",
     "\n",
-    "The EconML and DoWhy libraries complement each other in implementing this solution. On one hand, the DoWhy library can help [build a causal model, indentify the causal effect](#identify) and [test causal assumptions](#robustness). On the other hand, EconML’s `DML` based estimators can be used to take the discount variation in existing data, along with a rich set of user features, to [estimate heterogeneous price sensitivities](#estimate) that vary with multiple customer features. Then, the `SingleTreeCateInterpreter` provides a [presentation-ready summary](#interpret) of the key features that explain the biggest differences in responsiveness to a discount, and the `SingleTreePolicyInterpreter` recommends a [policy](#policy) on who should receive a discount in order to increase revenue (not only demand), which could help the company to set an optimal price for those users in the future."
+    "The EconML and DoWhy libraries complement each other in implementing this solution. On one hand, the DoWhy library can help [build a causal model, identify the causal effect](#identify) and [test causal assumptions](#robustness). On the other hand, EconML’s `DML` based estimators can be used to take the discount variation in existing data, along with a rich set of user features, to [estimate heterogeneous price sensitivities](#estimate) that vary with multiple customer features. Then, the `SingleTreeCateInterpreter` provides a [presentation-ready summary](#interpret) of the key features that explain the biggest differences in responsiveness to a discount, and the `SingleTreePolicyInterpreter` recommends a [policy](#policy) on who should receive a discount in order to increase revenue (not only demand), which could help the company to set an optimal price for those users in the future."
    ]
   },
   {
@@ -1007,7 +1007,7 @@
     "\\end{align}\n",
     "\n",
     "\n",
-    "With the decrease of price, revenue will increase only if $\\theta(X)+1<0$. Thus, we set `sample_treatment_cast=-1` here to learn **what kinds of customers we should give a small discount to maximum the revenue**.\n",
+    "With the decrease of price, revenue will increase only if $\\theta(X)+1<0$. Thus, we set `sample_treatment_costs=-1` here to learn **what kinds of customers we should give a small discount to maximum the revenue**.\n",
     "\n",
     "The EconML library includes policy interpretability tools such as `SingleTreePolicyInterpreter` that take in a treatment cost and the treatment effects to learn simple rules about which customers to target profitably. In the figure below we can see the model recommends to give discount for people with income less than $0.985$ and give original price for the others."
    ]