Top 20 Robotic Process Automation(RPA) Interview Questions and Answers

Ques: 1. What is Blue Prism’s Robotic Automation?

Answer: 

Robotic Automation infers process Automation’s where computer software drives present enterprise application software in a similar way that a user does. Automation is a gadget that operates other application software through the present application UI.

 

Ques: 2. What are benefits of Robotic Process Automation?

Answer: 

Benefits of RPA are:

• Faster: As bots are dealing with the execution here, a greater measure of work can be done in a relatively much shorter period. A faster delivery coupled with accuracy. 
• Consistency: It is a safe, non-invasive technology that doesn’t interfere with the inherent systems and provides impeccable consistency in performing the activities across the board, each time. 
• Cost Effective: It has been projected that using robotics cuts operational costs, Robots can operate 24*7 and take no leave, when compared to humans. 
• Increased Customer Satisfaction: Providing better quality of work with optimum accuracy and improved customer/client interaction leads to increased customer and client satisfaction. 
• Accuracy & Quality: RPA offers better services to processes that have a high probability of human error, thereby increasing accuracy. Robots are reliable, consistent and do not whine when expected to work tirelessly. 
• Improved Analytics: Having access to error free, perfect data from various sources would improve the quality of analytics in the process.

 

Ques: 3. What is the difference between thin client and thick client?

Answer:

Thin client: It is any application that we cannot get the quality properties while spying using any RPA tools.

e.g.  Any virtual environment

Thick client: It is any application that we get pretty handful of attribute features using RPA tools

e.g. calculator, Internet explorer.

 

Ques: 4. What are the important Phases of RPA Life Cycle?

Answer:

Phases of RPA Life Cycle:

• Analysis: The first phase in RPA begins with analysis. Business team and RPA Architect work together to understand a business process for RPA development. 
• Bot Development: RPA developer (Team) starts working on the requirement in their environment possibly a distinct development environment. 
• Testing: Some companies conduct Testing by Separate Testing Team, while some have a dedicated testing team which performs a dedicated QA like normal SDLC flow. Best Practice is to have a dedicated testing team which performs QA of developed bot. 
• Deployment and Maintenance: After the Development and Testing phases, a bot is ready for distribution and enters maintenance phase.

 

Ques: 5. What are Limitations of Robotic Process Automation?

Answer: 

Limitations of RPA are:

• RPA surely improves company efficiency by powering repetitive human effort, but there are limitations to the types of work that it can be applied to – especially ones that require judgment. 
• Enterprises need to be aware of various inputs coming from multiple sources. 
• It cannot read any data that is non-electronic with unstructured inputs. 
• RPA is not a cognitive computing solution. It cannot learn from experience and therefore has a ‘shelf life’. 
• Implementing RPA to a broken and incompetent process will not fix it. RPA is not a Business Process Management solution and does not bring an end-to-end process view.

 

Ques: 6. How is RPA going to impact the BPO offshore market?

Answer: 

All the large BPO providers have made bold statements about what RPA will do for their businesses. For instance, some declare that it will automate 50% of the FTEs performing the processes today. These are the companies that have stated targets and business plans to increase their revenues and are trying to move up the value chain. 

This is where we see a real gap in the market, which we are addressing by starting to offer services around Robotic BPO (R-BPO). If you have a process that is well-suited for automation, we can provide that as a service and handle the exceptions.

 

Ques: 7. What are the various categories of RPA tools?

Answer: 

All RPA tools can be categorized by the functionality they provide in these 3 dimensions:

1. Programming options: RPA bots need to be programmed and there are a few ways to program bots which involve trade-offs between complexity of bots and programming time. 
2. Cognitive capabilities: Programmed bots need to have cognitive capabilities to determine their actions based on inputs they gather from other systems. RPA tools provide a range of cognitive capabilities. 
3. Usage: Bots serve specific functions. Though most RPA tools can be used to build bots that serve all these functions, some tools are more optimized for attended or unattended automation. While unattended automation is batch-like background processes, in attended automation users, for example customer service reps, invoke bots like invoking macros.

 

Ques: 8. What’s the future of RPA?

Answer: 

There are some problems of RPA for which the leading solution providers are working to fix. All these solutions focus on the 2 most expensive portions of RPA deployment:

1. Design & development and
2. Maintenance.

There solutions are:

1. No code RPA: Enabling companies rely on cheaper resources and reduce RPA development time. 
2. Self-learning RPA: Automating process modelling using system logs and videos of users working on the process. 
3. Cognitive RPA: Enriching RPA with advanced functionality such as image processing and Natural Language Processing.

 

Ques: 9. What are reusable RPA plugins/bots?

Answer: 

Reusable RPA plugins/bots are programs that can be added to your RPA tool to take care of specific tasks like data extraction from invoices, manipulating dates in different databases, transcribing speech etc. Therefore, they reduce development efforts, error rates and implementation time.

 

RPA is a flexible automation platform. Therefore, rolling out RPA solutions require significant programming and customization. In this way, RPA is analogous to programming languages and platforms which are also flexible automation tools. Functions are critical in software development as they enable code reusability, reducing development time and errors. RPA is no different, reusability reduces RPA development times and programming errors.

 

Ques: 10. What are the common pitfalls need to be avoided in RPA implementation?

Answer: 

We have seen 3 types of pitfalls in RPA implementations:

1. Organizational pitfalls: Lack of commitment either from management or the team itself can delay any project and RPA projects are no exception.
2. Process pitfalls: Choosing an overly complex or insignificant process will lead to limited impact. For example, implementing RPA to an area like expense auditing where specialized solutions exist, can lead to significant effort without satisfying results.
3. Technical pitfalls: Choosing a difficult-to-use RPA tool can slow down development efforts.

 

Ques: 11. How the Robotic Process Automation is reliable and secure, and why RPA is significant?

Answer: 

For every business venture, the eventual goal in mind is to gain a bit more than has been invested in the first place as a capital. For that matter, RPA is a proven technology that reduces human labor by replacing it with robots that not tend to spawn prone error-ridden operational processes.

By propelling and encouraging parallel operations that range from fundamental front end and back end processes to the advanced cloud-based environments, RPA expedites the performance of the heaping workloads well before the deadline.

 

Ques: 12. Name the systems that can be robotically integrated with Blue Prism?

Answer: 

The notable distinctive feature of Blue Prism is its ability to incorporate diverse sources of technologies into its software systems. The integrated technologies in turn synchronized with Blue Prism software infrastructure, thus becoming robust and secure.

Rather than setting up individual adaptors for every single application in the software, Blue Prism comes packed with technologies and programmed with tools like Java, Windows, Green Screen, and Mainframe.

These settings are additionally allied with even more resolute tools so that they could be linked with Blue Prism. Without impacting the already existing systems, Blue Prism adapts with quite an adaptability the newly designed, built, and tested applications.

 

Ques: 13. What are the important stages of RPA lifecycle?

Answer: 

The various important stages of RPA lifecycle:

1). Analysis: To develop a potential RPA business process, RPA architects and business management team collaborate to analyse and identify the business process.

2). Bot Development: At this phase, an agent that simulate the human activity is induced so that the robotic process could develop.

3). Testing: The developed bot is tested with the specially designed QA that certifies the successful architecture of the product.

4). Deployment and Maintenance: Once an RPA product is done, it is launched to be deployed to the user-end and maintained with carefully developed additional tool systems.

 

Ques: 14. How Robotic Automation is different from macros and screen scratching?

Answer: 

Unlike the generation-old applications like screen scratching and macros, any given application that is used by human can also be used by robots. RPA is equipped with handling complex applications like web frameworks, mainframes, web service apps, and can work efficiently with Application Programming Interface (API) hosting services.

These applications are read by the robot, either through existing APIs where they are prolonging, through the operating systems before applications appear. In this case, the modern robots reads an application screen in context and in the same way a user does. As part of the robot training, it is shown how to read the display of the application much like a user.

 

Ques: 15. What are the framework types used in Automation Anywhere?

Answer: 

The following frameworks are used in Automation Anywhere:

1. Data-driven automation framework.
2. Keyword Driven Automation Framework.
3. Modular automation framework.
4. Hybrid Automation Framework.

 

Ques: 16. What is the role of the RPA developer?

Answer: 

Process Designer is responsible for understanding the current process. He / she ensures that the people working on the RPA project are synchronized. It also monitors the changes that occur after the implementation of the feedback during the development or test phase, keeping project specifications intact.

 

Ques:17. Does RPA store data?

Answer: 

The RPA stores data Although all are known as RPA, each of them is selected according to the processes or tasks that the organization wants the robots to handle. There are ‘probots’, which process data, ‘knowbots’ to collect and store data, and ‘chatbots’ that act as virtual agents to respond to customer queries in real time.

 

Ques: 18. What can the RPA not do?

Answer: 

Of robotic processes, also known as RPA, is a rapid and important change that is invading many industries. … RPA can help your company employees set up computer software or a robot to capture and interpret existing applications to help manufacture, transfigure, and analyse data.

 

Ques:19. What are the various features of RPA?

Answer: 

The important features of RPA are:

1). User-Friendly: RPA selection starts inside business tasks rather inside IT divisions. RPA ventures require less IT aptitudes and less speculation. In the long run, the robotization is brought down at a generous rate.

2). Rick Free: RPA (Robotic Process automation) is low complexity and risk-free from other Tools. RPA access to end users’ systems Through a controlled user interface, hence Increasing the Important of underlying systems programming.

3). Code Free: RPA (Robotic Process automation) doesn’t require programming skills. anyone can Learn RPA With Simple Effect Because Its run’s without coding, with any subject expertise, can be trained to automate RPA tools instantly. RPA tool Designed with charts and flowcharts.

 

Ques: 20. Is Blue Prism’s Robotic Automation Platform secure and auditable?

Answer: 

Security and auditability are consolidated into the Blue Prism robotic automation platform at various levels. The runtime environment is totally separate to the process editing environment.

Approvals to design, create, edit and run processes and business objects are specific to each authorized user. A full audit trail of changes to any process is kept, and comparisons of the before and after effect of changes are provided.

The log created at run-time for each process provides a detailed, time-stamped history of every action and decision taken within an automated process. Our clients tend to find that running a process with Blue Prism gives them a lot more control than a manual process, and from a compliance point of view assures that processes are run consistently, in line with the process definition.

Top 20 Machine Learning Interview Questions and Answers

Ques: 1. What is the difference between supervised and unsupervised machine learning?

Answer:

Supervised learning requires training labelled data. For example, in order to do classification (a supervised learning task), you’ll need to first label the data you’ll use to train the model to classify data into your labelled groups. Unsupervised learning, in contrast, does not require labelling data explicitly.

 

Ques: 2. What is Overfitting? And how do you ensure you’re not overfitting with a model?

Answer:

Over-fitting occurs when a model studies the training data to such an extent that it negatively influences the performance of the model on new data. This means that the disturbance in the training data is recorded and learned as concepts by the model. But the problem here is that these concepts do not apply to the testing data and negatively impact the model’s ability to classify the new data, hence reducing the accuracy on the testing data.

Collect more data so that the model can be trained with varied samples. Use assembling methods, such as Random Forest. It is based on the idea of bagging, which is used to reduce the variation in the predictions by combining the result of multiple Decision trees on different samples of the data set.

 

Ques: 3. What do you understand by precision and recall?

Answer:

Recall is also known as the true positive rate: the number of positives your model claims compared to the actual number of positives there are throughout the data. Precision is also known as the positive predictive value, and it is a measure of the number of accurate positives your model claims compared to the number of positives it actually claims. It can be easier to think of recall and precision in the context of a case where you’ve predicted that there were 10 apples and 5 oranges in a case of 10 apples. You’d have perfect recall (there are actually 10 apples, and you predicted there would be 10) but 66.7% precision because out of the 15 events you predicted, only 10 (the apples) are correct.

 

Ques: 4. What are collinearity and multi collinearity?

Answer:

Collinearity occurs when two predictor variables (e.g., x1 and x2) in a multiple regression have some correlation.

Multi collinearity occurs when more than two predictor variables (e.g., x1, x2, and x3) are inter-correlated.

 

Ques: 5. What’s the difference between Type I and Type II error?

Answer:

Don’t think that this is a trick question! Many machine learning interview questions will be an attempt to lob basic questions at you just to make sure you’re on top of your game and you’ve prepared all of your bases.

Type I error is a false positive, while Type II error is a false negative. Briefly stated, Type I error means claiming something has happened when it hasn’t, while Type II error means that you claim nothing is happening when in fact something is.

A clever way to think about this is to think of Type I error as telling a man he is pregnant, while Type II error means you tell a pregnant woman she isn’t carrying a baby.

 

Ques: 6. What is A/B Testing?

Answer:

A/B is Statistical hypothesis testing for randomized experiment with two variables A and B. It is used to compare two models that use different predictor variables in order to check which variable fits best for a given sample of data.

Consider a scenario where you’ve created two models (using different predictor variables) that can be used to recommend products for an e-commerce platform.

A/B Testing can be used to compare these two models to check which one best recommends products to a customer.

 

Ques: 7. What is deep learning, and how does it contrast with other machine learning algorithms?

Answer:

Deep learning is a subset of machine learning that is concerned with neural networks: how to use back propagation and certain principles from neuroscience to more accurately model large sets of unlabelled or semi-structured data. In that sense, deep learning represents an unsupervised learning algorithm that learns representations of data through the use of neural nets.

 

Ques: 8. Name a few libraries in Python used for Data Analysis and Scientific Computations.

Answer:

Here is a list of Python libraries mainly used for Data Analysis: 

• NumPy
• SciPy
• Pandas 
• SciKit
• Matplotlib 
• Seaborn 
• Bokeh

 

Ques: 9. Which is more important to you– model accuracy, or model performance?

Answer:

This question tests your grasp of the nuances of machine learning model performance! Machine learning interview questions often look towards the details. There are models with higher accuracy that can perform worse in predictive power — how does that make sense?

Well, it has everything to do with how model accuracy is only a subset of model performance, and at that, a sometimes misleading one. For example, if you wanted to detect fraud in a massive data set with a sample of millions, a more accurate model would most likely predict no fraud at all if only a vast minority of cases were fraud. However, this would be useless for a predictive model — a model designed to find fraud that asserted there was no fraud at all! Questions like this help you demonstrate that you understand model accuracy isn’t the be-all and end-all of model performance.

 

Ques: 10. How are NumPy and SciPy related?

Answer:

NumPy is part of SciPy. NumPy defines arrays along with some basic numerical functions like indexing, sorting, reshaping, etc.

SciPy implements computations such as numerical integration, optimization and machine learning using NumPy’s functionality.

 

Ques: 11. How would you handle an imbalanced dataset?

Answer:

An imbalanced dataset is when you have, for example, a classification test and 90% of the data is in one class. That leads to problems: an accuracy of 90% can be skewed if you have no predictive power on the other category of data! Here are a few tactics to get over the hump:

1. Collect more data to even the imbalances in the dataset.
2. Re-sample the dataset to correct for imbalances.
3. Try a different algorithm altogether on your dataset.

What’s important here is that you have a keen sense for what damage an unbalanced dataset can cause, and how to balance that.

 

Ques: 12: Is rotation necessary in PCA? If yes, Why? What will happen if you don’t rotate the components?

Answer:

Yes, rotation (orthogonal) is necessary because it maximizes the difference between variance captured by the component. This makes the components easier to interpret. Not to forget, that’s the motive of doing PCA where, we aim to select fewer components (than features) which can explain the maximum variance in the data set. By doing rotation, the relative location of the components doesn’t change, it only changes the actual coordinates of the points.

If we don’t rotate the components, the effect of PCA will diminish and we’ll have to select more number of components to explain variance in the data set.

 

Ques: 13. What’s the “kernel trick” and how is it useful?

Answer:

The Kernel trick involves kernel functions that can enable in higher-dimension spaces without explicitly calculating the coordinates of points within that dimension: instead, kernel functions compute the inner products between the images of all pairs of data in a feature space. This allows them the very useful attribute of calculating the coordinates of higher dimensions while being computationally cheaper than the explicit calculation of said coordinates. Many algorithms can be expressed in terms of inner products. Using the kernel trick enables us effectively run algorithms in a high-dimensional space with lower-dimensional data.

 

Ques: 14. Explain prior probability, likelihood and marginal likelihood in context of naiveBayes algorithm?

Answer:

Prior probability is nothing but, the proportion of dependent (binary) variable in the data set. It is the closest guess you can make about a class, without any further information. For example: In a data set, the dependent variable is binary (1 and 0). The proportion of 1 (spam) is 70% and 0 (not spam) is 30%. Hence, we can estimate that there are 70% chances that any new email would  be classified as spam.

Likelihood is the probability of classifying a given observation as 1 in presence of some other variable. For example: The probability that the word ‘FREE’ is used in previous spam message is likelihood. Marginal likelihood is, the probability that the word ‘FREE’ is used in any message.

 

Ques: 15. Do you have experience with Spark or big data tools for machine learning?

Answer:

You’ll want to get familiar with the meaning of big data for different companies and the different tools they’ll want. Spark is the big data tool most in demand now, able to handle immense datasets with speed. Be honest if you don’t have experience with the tools demanded, but also take a look at job descriptions and see what tools pop up: you’ll want to invest in familiarizing yourself with them.

 

Ques: 16: You came to know that your model is suffering from low bias and high variance. Which algorithm should you use to tackle it? Why?

Answer:

Low bias occurs when the model’s predicted values are near to actual values. In other words, the model becomes flexible enough to mimic the training data distribution. While it sounds like great achievement, but not to forget, a flexible model has no generalization capabilities. It means, when this model is tested on an unseen data, it gives disappointing results.

In such situations, we can use bagging algorithm (like random forest) to tackle high variance problem. Bagging algorithms divides a data set into subsets made with repeated randomized sampling. Then, these samples are used to generate  a set of models using a single learning algorithm. Later, the model predictions are combined using voting (classification) or averaging (regression).

Also, to combat high variance, we can:

Use regularization technique, where higher model coefficients get penalized, hence lowering model complexity.

Use top n features from variable importance chart. May be, with all the variable in the data set, the algorithm is having difficulty in finding the meaningful signal.

 

Ques 17. Which data visualization libraries do you use? What are your thoughts on the best data visualization tools?

Answer:

What’s important here is to define your views on how to properly visualize data and your personal preferences when it comes to tools. Popular tools include R’s ggplot, Python’s seaborn and matplotlib, and tools such as Plot.ly and Tableau.

 

Ques: 18. How is kNN different from kmeans clustering?

Answer: 

Don’t get mislead by ‘k’ in their names. You should know that the fundamental difference between both these algorithms is,

• kmeans is unsupervised in nature and kNN is supervised in nature.
• kmeans is a clustering algorithm. kNN is a classification (or regression) algorithm.
• kmeans algorithm partitions a data set into clusters such that a cluster formed is homogeneous and the points in each cluster are close to each other.

The algorithm tries to maintain enough separability between these clusters. Due to unsupervised nature, the clusters have no labels.

KNN algorithm tries to classify an unlabelled observation based on its k (can be any number ) surrounding neighbors. It is also known as lazy learner because it involves minimal training of model. Hence, it doesn’t use training data to make generalization on unseen data set.

 

Ques: 19. Is it better to have too many false positives or too many false negatives? Explain.

Answer:

It depends on the question as well as on the domain for which we are trying to solve the problem. If you’re using Machine Learning in the domain of medical testing, then a false negative is very risky, since the report will not show any health problem when a person is actually unwell. Similarly, if Machine Learning is used in spam detection, then a false positive is very risky because the algorithm may classify an important email as spam.

 

Ques: 20. What is the difference between Gini Impurity and Entropy in a Decision Tree?

Answer:

Gini Impurity and Entropy are the metrics used for deciding how to split a Decision Tree.

Gini measurement is the probability of a random sample being classified correctly if you randomly pick a label according to the distribution in the branch.

Entropy is a measurement to calculate the lack of information. You calculate the Information Gain (difference in entropies) by making a split. This measure helps to reduce the uncertainty about the output label.