Artificial Intelligence, Robotic Process Automation and Machine Learning

RPA (Robotic process automation)

1.    RPA is the process of integrating all the business operations with the help of robots to reduce human interference.

2.    It is the wide application of the technology controlled by the business logic. It is used to speed up the entire business process.

3.    It automates the business which is rule-based and structured.

4.    It uses ‘machine learning’ and ‘artificial intelligence’ capabilities to do the bigger volume workload, which was previously done manually.

5.    All the tasks to be automated here are queries and record.

6.    RPA is included with the system bots. They can work as a human worker.

7.    RPA bots can handle the entire application.

8.    RPA bots perform repetitive tasks much faster. This allows employees to perform higher-value work.

9.    RPA bot can be programmed to finish any repetitive task. This provides flexibility.

10.    RPA bots are easy to integrate. They can take the input and they can analyse the output of all the windows applications.

11.    RPA bots are divided into three categories i.e. Probots, knowbots, and chatbots.

12.    Probots follow the simple rules to process the data.

13.    Knowbots search the user-specified information through the internet.

14.    Chatbots are the virtual assistants. They keep on responding to the customer queries.

15.    RPA reduces the labour costs of the organization.

16.    RPA software adopts the newly changing environment.

17.    RPA is the scalable technology that can be applied to all the industrial departments. It can be applied to the work which is more prone to manual error, manual data entry, and high-volume work.

18.    RPA are classified as Attended automation, Unattended automation, and Hybrid RPA.

19.    Attended automation tools require manual interference to finish the process.

20.    Unattended automation tools have self-decision-making capabilities.

21.    Hybrid RPA is the combination of Attended and Unattended automation.

22.    Decision making is improved with the implementation of RPA bots.

23.    Recent RPA tools used are UiPath, Blue Prism, Automation anywhere.

24.    RPA has less technical barriers. It is basically code-free technology; hence anyone can use it.

25.    RPA increases work accuracy.

26.    RPA bots work 24*7.

27.    It is used in collecting social media statistics.

28.    It is also in extracting structured data from documents, payroll, new recruitment hiring.

29.    It is also used in the health industry for patient registration and billing.

30.    In the insurance industry, it is used for claim processing.

31.    In manufacturing, it is used to create the bill of materials and calculate the total sales, inventory management, and purchase order management.

32.    In travel and logistics, it used for the ticket booking, and passenger details.

33.    In the banking sector, it is used for card activation and fraud claims.

34.    In BPO, it automates all the manual operations and it gives the best customer service.

35.    It is also used to collect the data from different web portals.

36.    In the telecom sector, it is used to generate the bills.

37.    Any industry can use the RPA to increase its productivity, scale, and accuracy. Its costs are recovered in 4 years.

 

Artificial Intelligence (AI)

1.    It is the ability of the machine to perform all the tasks performed as like the human beings.

2.    AI are the programmed devices and software’s that behave as humans do.

3.    AI prepares the programmed device or software for the specified task and allows it to improve on its own.

4.    AI are trained programmes capable to perform and work independently.

5.    AI are classified as Narrow AI, Artificial general intelligence (AGI), and Artificial superintelligence (ASI)

6.    Narrow AI is sometimes called as weak AI. It performs the single tasks and it operates under the limited context. Ex. Alexa, Google Home, Siri, self-driving cars.

7.    ASI is the hypothetical AI. It is the way in which all the programmed devices of AI behave and think independently, and they suppress human intelligence.

8.    AGI are the strong AI, seen in the movies. Ex. Humanoid.

9.    AI is tested with the Turing Test, Cognitive Modelling Approach, Law of Thought Approach and Rational Agent Approach.

10.    Turing test finds out the artificial intelligence entity is effectively holding the conversation with the human agent.

11.    Cognitive Modelling Approach builds the AI model based on the human understanding.

12.    Law of Thought Approach applies the logical statements to the AI software’s.

13.    Rational agent tries to achieve the best possible outcome in the circumstances. It acts as a dynamic agent.

14.    Types of AI are reactive machines, limited memory, a theory of mind, and self-awareness.

15.    Reactive machines have the capacity of not using the past memory for the current situation decisions.

16.    Limited memory machine has the capacity to use the past memory to perform the current task. Their memory is a short term.Ex. Self-driven cars, traffic lights etc.

17.    Theory of mind has the capacity to react as per the people emotions and thoughts. It can adjust with the people around.

18.    Self-awareness machines are more aware of their needs and their senses than humans.

19.    AI enabled virtual assistant is widely used in the hospitals for the patient visits.

20.    AI applications are Machine learning, Deep learning, Neural Networks, Evolutionary Algorithm, Cognitive Computing, Natural Language Processing, Computer vision.

21.    Machine learning is the application that provides the computer systems with the ability to learn automatically and learn from the experience without being programmed. Machine learning requires a lot of data.

22.    Machine learning focuses on the development of the algorithms that analyse the given data and make the prescribed predictions. Ex. Movie recommendations in the Netflix.

23.    Deep learning uses many neutral networks for many layers of processing networks. Ex. Speech recognition.

24.    Neutral Networks try to use the human brain approach to analyze the data. It finds out the wide data for analysis that is much complex for the human brain.

25.    Natural language processing (NLP) allows the computer to recognize and produce the human speech and language. The goal of NLP is to have effective interactions between the man and the machine for the daily activities. Ex. Google translator.

26.    Computer vision tries to understand the image and pictures with the PDF documents by breaking it into the different parts.

27.    The graphical processing unit is required for AI to carry out the millions of data.

28.    The best example of AI is Alexa and google home.

29.    AI is used to plan the travel, trips and the stays with the help of virtual AI assistant.

30.    AI applications are used to provide the personalized X-ray copy and the medicines.

31.    Stock management is improved with the help of AI.

32.    AI techniques are used to identify fraudulent transactions in the banking sector.

33.    AI performs high volume computerized tasks easily.

34.    Recent examples are Amazon Alexa, Google Home, Apple Siri, Tesla, Cogito.

 

Machine Learning (ML)

1.    Machine learning is the application of “Artificial Intelligence’’ which provides a system ability to learn, think, react and work automatically and improve from its experience without any manual programming.

2.    ML focuses on the programmes that can access the data and use it to learn for themselves.

3.   The basic motive of the ML is to allow the computer programmes to learn and work themselves without any external output.

4.    ML is the statistics that make the predictions using the computers.

5.  Machine learning methods are classified as Supervised Machine learning, Unsupervised Machine learning, Semi-supervised machine learning, Reinforcement machine learning.

6.    Supervised Machine learning is the learning in which training is given to the machine with the help of data which is tagged with the correct answer. Ex. Basket is filled with different types of fruits. The first step is to teach the machine the different shapes of fruits along with the colour one by one.

7.    Unsupervised learning is the training of the machine using the information that is never classified and it allows the algorithm to act on it.

8.    Machine gathers the unsorted information according to the similarities without any training. In this case, no training is given to the machine.

9.    Semi- supervised machine learning is a combination of supervised and unsupervised machine learning.

10.    Reinforcement machine learning is the technique which helps the systems to learn by trial and error method.

11.    Machine learning requires careful preparation of lots of data.

12.    ML is used to find out which product you are looking to buy on Amazon.

13.    ML algorithm consists of Representation, Evaluation and Optimization.

14.    Recent example of ML is FB News Feed.

15.    Other example of ML in Govt service is to find the fraud in the banking sector. Another ex is Image recognition, Speech recognition, Medical diagnosis etc.

16.     Another ex.GPS services, Video navigation, Email Spam, and malware filtering.

 

 

RPA	AI	ML
Process driven	Data driven	Data driven
It is a software robot.	It is a simulation of human intelligence.	It is the application of AI.
Automation of high-volume repetitive tasks	It performs all the operations as like the humans do.	Ability of the machine to work and learn independently without any external support.
It is used to reduce human interference	It is used with the help of human commands. (It can work on its own)	It does not require any external interference.
Good for clerical work	Good for daily work.	Good for daily and professional work.
Coding is not required.	Slightly coding is needed.	Slightly coding is needed.
Ex. Applications in BPO, bank, and inventory management.	Ex. Alexa, Siri, Tesla.	Ex. FB news feed.