You are on page 1of 48
www.askbooks.net *AKTU Quantums «Toppers Notes *Books *Practical Files Projects *IITJEE Books www.askbooks.net All AKTU QUANTUMS are available _akbooks.net does not own the materia neither created it nor scanned it. We provide the links tothe materials which re already availabe onthe internet. * Your complete engineering solution. * Hub of educational books eee ee Tee RU rl CT SAUL. DOE on this website are submitted by readers you can also donate ebooks/study materials. 2. We don't intend to infringe any copyrighted material. Pee are ura cue kak acd website you can kindly report us, we will remove it asap. waa Cokie Con PART-1 ‘The Evolution of Al tothe Present. An Overview to AI Que tats | What do you mean by artificial intelligence ? Define its goals. Part-l + The Evolution of Al to 1. Antifcial Intelligence (AD)is an area of computer science that emphasizes Ths Brohate conn D2ME 9 15M | inc ereation oftlligent machines that work and react lke Burns ae. | 2 Athas become an essential part of the technology industry: mart2 + Various Approaches to AL BM to 1-094 41 Research assvited with artificial intelligence is hihly technical and © Part3. What should All Eagincers Thoculized The core probleme of artificial intelligence include now about AU? 1-99 to 1-1aMt Programming computer or certain trait sch as i Knowle Paria: 4 + Other Emerging Technologies... 1-12M to 1-14M ii Reasoning + Al and Bihical Concerns .. oe AA14M to 117M iii Problem solving in. Perception ee Lan vi Planing vit Ability to manipulate and move objects Goals of A 1. To create expert systems : The systems which exhibit intelligent behaviour, learn, demonstrate, explain, and advice its users. 2 ‘Toimplement human intelligence in machines : Creating eystems ‘that understand, think, earn, and behave like humans QuetZ] What are the different branches of artificial Intelligence? Discuss some of the branches and progress made in their field 11M (Sem-1 & 2) 13M Bem g 9) ieee Dilferent ranches 1. Machine Learning (ML) Machine Vetted were th target defined and the steps Mr learnaby the machine itselfby training, "8 Poreranpletoidentify asimple object such as an apple oo Bole outed rig mate ictures of object Bore phe machine fo define the steps aden ‘an apple or an orange. entitle 2 Natural Language Processing (NLP) i. NLPisdefined as the automatic manipulation of naturs like speech and text, by software. ural language, 4 Forezanple, email spam detection which has improw Frenne nas improved the mail Vision Machin visa captre and analyses visual in information a camera, analog-to-digital conversion, and digital si roe ae as tal signal processing 1 Robotic is ld of engineering focused on - manufacturing foots. ve Seale cece {osu ts thar die fr aman rm i Farexamle carer ies nhospitals office clean foods and preparing foods nthtels eve, SLE eaniine Te Be |] Define therole of the machine intelligence in the human iAnewer ey. idiot oe ef the para 2 Redsbeisernmanetieapre tine > betta maint ‘In bumaa life, mact fh ‘tebe chine learning solves many problems of daily purpost came ee anit require intelligence such as complex fentertyaanceyane eee oes Se mst ‘role in following areas: ren ee ese Beans to acquire new things from the set of f or experiences. It refers to the change in subject's behaviour toa give that stuations ©” t4stion brought by repeated experiences 8 of Aland progress made in these fields ; ‘An Overview to AI jm (Sem-1 & 2) from given facts ii, Reasoning : Reasoning means to infer facts Tienances re clasifed as either deductive or inductive and the reasoning isto draw inferences appropriate tothe situation. move towards the fil, Problem solving To solve problem means 9 era lcthin tnt ofubsar defined and o gels aso defined which Stabe ahievedby wing these ues ive Language understanding : It means to under Teanmagemeaning, language isa yom of ssh Uireonvention. the mesning-by-convention Preonse and is very diferent rom natural meaning GaodAs | Describe briefly the evolution of artificial intelligence- Evolution of AI: 1. Beginning of AI (1943): The concept of pot limited to the computer sciences disciplines, various other areas, 2 Alkcnowledge-based expert system (1970) rule-based system to capture knowledge ia t Statements or as deision ree. |g. Machine learning (1998) :There are two types of machine learning Formal : The formal type of machine learning is a computer vreneam that learns from experience in respect to some task and [lcreases performance based on that experience. Informal : The informal involves giving computers the ability to Tearn without explicily programming the capability. 4. Supervised learning (2004) : The supervised learning is based on Stn the corect answers and having the eompater mapping inputs to tutputs. For example, Spam filters: Softwares rained to learn and distinguish between Shim and non-spam messages (For example, e-mail filters) ji. Facial recognition: It is used by cameras to focus and via photo ting sofware to tag persons (For example, Facebook). & Unsupervised learning (2010) : Unsupervised learning is the reverse sremnpevioed learning where the correct answers are unknown, For example, ‘Clustering algorithm : Used to find patterns in datasets and hen roup that date into different coherent clusters. Market segmentation : Targeting customers based on regions, Men dislikes, wher the consumer makes purchases, te. This it considered targeted marketing. stand natural wing meaning distinctive of JfAI began around 1943. AT is ‘bat can be seen in ) An AT system often uses the form of ifthen ere Artificial Intelligence For Engineering 15M (Sem-1 @ 9) & Genetic programming (2010) : Genetic programming is an cer evotutionary proces toimprove algorithms, ‘1. Futureof Al (2019onwards): There are many challenges in mis Fane etalgence Humans acquire common senses that are inti 8 Hatin te reason ratioally. For example, the colour of a blue e"® blue. PAR Various Approaches to-Al. idea that Questions-Answers Long Answer Type and Medium Answer Type Questions Que 15: ] What are the different approaches to AI? Answer ifferent approaches to Al: Artificial Intelligence AD, as abroad field, encompasses many different approaches ranging from top-down knowl epee ranging tem tpdown knowledge representation fo "Thre are thre relied Te fed concepts that have been frequently sed in ars: Al machine learting and depleering 3 Ingeneral Ait trod dtand dopleumag ranars le concep, machin learning isa subfield in special ype of machine learning. lustre th ations among these thre concepts. 1-6M (Sem-1& 2) 5 A ‘An Overview to AI ‘While the broad field of AL includes many approaches, its popularity is lorgely due to the outstanding performances of machine earning and deep learning, Machine learning i. Machine learning is an application of Artificial Intelligence (AD that provides systems the ability toautomatically learn and improve from experience without being explicitly programmed. Machine learning focuses on the development of computer programs that ean access data. ‘The primary aim isto allow the computers to learn automatically without human intervention or assistance and adjust actions ‘accordingly. iv. Machine learning enables analysis of massive quantities of data. Deeplearning: i. Deep learnings the subfield of artif ‘on reating largeneural network mod faceurate data-driven decisions. fi Deep learning is used where the data is complex datasets fii Facebook uses deep leari ‘conversations. jv. Google and Mierosoft all use deep learning for image search and machine translation. ficial intelligence that focuses jels that are capable of making and has large ning to analyze text in online Quol6. | What are the advantages and disadvantages of machine earning ? Advantages of machine learning are 1 Easily identifies trends and patterns : ‘Machine learning ean review large volumes of data and discover specific trends and patterns that would not be apparent to humans. b Foran.e-commerce website like Flipkart, it serves to understand the browsing behaviour and purchase histories ofits users to help cater to the ight products, deals, and reminders relevant to them, c. _ It uses the results toreveal relevant advertisements to them. No human intervention needed (automation) : Machine learning ‘does not require physical force i.., no human intervention is needed. Continuous improvement : ‘a. MLalgorithms gain experience; they keep improving in accuracy and efficiency. ha yd “Arifiil Intelligence For Engineering 17M Sem-1a 9 + Asthe amount of data keeps growirg, algorithms learn securate predictions faster. 10 make 4. Handling multi-dimensional and multi-variety data ; Machine leering lorthms ae god at handling data that any ‘multidimensional and multicvariely, and they ean do tnt ddynamicor uncertain environments, do this in Disadvantages of machine learning are : 1, Data acquisition : ‘4 Machine learning requires massive data sets to train on, any ‘Should beinclusivefunbiased and ofgood quality.” "4 ee 2 Time and resources 4. ML needs enough in oat the algorithms learn and trough to fall tr purpose with a coneiderbte sceuracyand relevancy. 1k Tealroneeds massive resources to fneton {Interpretation of results ‘To accurately interpret retult generated bythe agri mot earful chore th alors for our purpose mt N° 4. High errorusceptbility 4 Machine ering inautonomousbuthighly susceptible to errors, b. Ittakes time to, ' recognize the source ofthe issue, tocorrect it the issue, Qe 17:5] Differentiate between machine learning and dee ‘ tween machine lea} ine learning > develop mount of and even longer sewer S.No] Maca 1. | ss Deep Learning Works on salt amo {stn acanameunt af | Works on large amount of 2 [Dependent = | machine. on low-end | Heavily dependent on high end machin 3 [Dividesthetasken e. Ives into sub-tasks, Ives, ales them individually ang | Sv°S Problem end to end. |__| finatiy combine the osulys, 4 [S[teteres tine tran lessee | NESS Takes more time to tral — time to test the data, Less time to test the data. 1-8M (Sem-1&2) ‘An Overview to Al aeTS,| What are the applications of deep learning ? 1, Automatic text gene ‘a. Corpusof text is learned and from this model new text is generated, word: by-word, character-by-character. b. Then this model is capable of learning how to spell, punctuate, form sentences, or it may even eapture the style. 2. Healthcare : Helps in diagnosing various diseases and treating it a. Automatic machine translation : Certain words, sentences or ‘Phrases in one language is transformed into another language. 4. Image recognition : Recognizes and identifies peoples and objects in {mages as well as to understand content and context. This area is tlready being used in gaming, retail, tourism, et. 1 Predictingearthquakes : Teaches a computer to perform viscoelastic ‘computations which are used in predicting earthquakes. (ons of artificial intelligence. newer Applications of artificial intelligence : 1. Gaming: At plays crucial role in strategie games such as chess, poker, tietac-oe, ete, where machine ean think of large number of possible positions based on heuristic knowledge. 2 Natural language processing : It is possible to interact with the ‘Computer that understands natural language spoken by humans. 8. Expert systems: There are some applications which integrate machine, oftware, and special information to impart reasoning and advising, ‘They provide explanation and advice to the users. 4. Vision systems: hese systems understand, interpret, and comprehend ‘visual input on the computer. Speech recognition :Some intelligent systems are capable of hearing ‘and comprehending the language in terms of sentences and their meanings while a human talls to it It ean handle different accents, “lang words, noise inthe background, change in human's noise due to cold te & Handwriting recognition : The handwriting recognition software reads the text written on paper by a pen or on screen by a stylus. Itean ecognize the shapes ofthe letters and convert it into editable text. ‘Artificial Intelligence For Engineering 1 1-9M (Sem-1& 2) ‘intelligent robots : Robots are able to perform the tasks given by a yhuman, They have sensors to detect physical data from the real world, ‘They have efficient processors, multiple sensors and huge memory, to exhibit intlligenee 3 ~ [PART-3 What should All Engineers Know About AI? 7 Qucstions-Anewers ‘Loni Answer Type and Medium Answer Type Questions Que iA0,] What engineers need to know about Artificial Intelligence? L ‘Artificial intelligence (AD) systems by their nature are software intensive. To create viable and trusted AI systoms, engineers need technologies and standards. Following arethe key aspects and elements of Al that engineers must ‘understand to work with emerging systems A. Introductory Concopts: AI, ML, and Deop Learning : 1. Alis define as _- tha bility of machine to perform task that normaly require iumanintelgence-for example, revoglcingptiora, learning from expert, drawing solaonm, ming prediction, of taking aeton- wheter digital or athe amar software behind tstonomove pha syateme 2 Machi arin ML) a part ofA is define a: Aotom hat lear and improvesits performance at some by using data and experience. be a ot 2 Deeplaring i define aa family of machine learni ine learning techniques whose models extract important features by iteratively transforming the data, “goin deeper toward meant ever” tard meaning pattern in the dataset with sch 1-10M(Sem-1& 2) ‘An Overview to AI ‘4. Unlike traditional machine learning methods, in which the ereator ‘of the model has to choose and encode features ahead of time, deep learning enables a model to at ratomatically learn features that matter. B. ATEn sering Concepls: Al depends on the human element : L ‘Al augments, but does not replace, human knowledge and ‘expertise. ‘This basic understanding affects engineers of AT systems in two dimensions : human-machine teaming and the probabilistic nature of AI “answers.” Engineers developing AI systems must account for the interactions between the system and the people who build and use it (human-machine teaming). Often, the succecs of those interactions comes down to trust ‘and transparency. Further, AI will produce probabilistic answers. How does the engineers of AI systems know when a prediction is bad? [AT depends on labeled and unlabeled data as well as the ystems that store and access it : 1 |. amounts of datato 5 L ‘The development in Al is due to the availability of data and the speed at which today’s computers can process it. Al systems can classify, categorize, and partition massive jake the relevant information available for humans to analyze and make decisions. Engineers must consider the data and the hardware and software aystoms that support that data, Large amounts of data require a computing environment ‘that has the capacity to handle it. ‘Managing data requires designing storage solutions around physical data constraints and typos of queries desired, One AI, many algorithms : When we talk about AI, ML, and deep learning, we are referring to many different algorithms, many different ‘approaches, not all of which are neural-network based. ‘Artificial Intelligence For Engineering 111M (Sem-1 & 2) 2 Many of the algorithms used in AI were generated in the 1950s, 1960s, oF 1970s. 8. For example, the A shortest path algorithm was conceived, {in the 1950s, and improved on in the 1960s. iv, ‘The insight is the benefit of AT 1, Engineers knows that it is impossible to test a system in every situation it will ever encounter. 2. An Al system can find an answer to never-seen-before situations that is insightful and has a very good probability of being correct. 3, However tie not necessarily correct, but probabil 4. Thus, gainingincreased confidence in Als hard for engineers ‘who need to focus on creating and validating a system. ¥. Ar Al system depends on the system under which it runs: 1, When building a system that doesnot incorporate Al, you can build and test tin icoation 2 Then deploy it and be certain itis going to behave just asit did inthe lab. 3. An ATsystem depends on the conditions under which the AI ‘Fans and what the AI system is sensing, and this context adds another level of complexity. i] What are the responsibilities of an AI Engineer ? ‘ofan AI Engineer : Convert the machine learning models inte application program interfaces (APIs) so that other applications ean use it Build Al models from seratch and help the different components of the organization (such as product managers and stakeholders) understand what results they gain from the model Build dats ingestion and data transformation infrastructure, ‘Automate infrastructure that the data science team uses. Pes tite nals and tine organization canmalebeterinfomeddecony nuns‘ be oes Setup ant manage Al development and product infrastructure 112M Gem-1 & 2) 1 — ‘An Overview to AI Be a good team player. Emerging technologies in artificial intelligence are i Al-enhanced Analytics Solutions : i. This category helps to organize the customer journey and experience. i Those systems can understand the customer, learn preferences, predict next best action/solution, and surface insights. iil, This is atop priority area for the contact center az Al-enhanced “analytes solutions can deliver now and stronger business benefit. Deep Learning (DL) i. DLisatypeofmachine learning algorithm that has the ability to ‘generate better predictions/insights, seale upwith largedata sets, ‘and reduce the effort to build the model. i, Inthe contact contor, DLs used in conversational systems (speech recording, NLG, NLU, ete), Speech Analytics, and other arcas, Natural Language Generation (NLG) : NLGis a part of the tech stack in conversational systems. NUG uses advanced Al algorithms to generate speech from text. i, NLGisused to gonerate speech in Alexa, in Virtual Assistants and ina Natural Language IVR. iv. NG is used as part of the SmartCare conversational platform ‘that powers IVR and Chatbot channels Artificial Intelligence For Engineering 119M (Sem-1& 2) 4. Speech Analytics : i. Speech analytics uses AI technology to recognize speech, convert Speech int text, and perform analytics on the text data et i This technology is used in many contact centers to improve customer interactions and agent performance. 5 Al foreyber-security applications: Artificial intelligerce and machine learning technology is Increasingly finding ts way into cybersecurity systems for both corporate systems and home security. ii AU and machine learning technology can be employed to help ‘dentiy threat, including variants of earlier threats. [AL-powered eybersectrtytols can collect data from a company’s transactional systems, communications networks, digital activity and websites, as wells from external pubic sources, and utilize ‘AI algorithms to recognize patterns and identify threatening activity je. Im home security AL system is integrated with consumer video cameras and intruder alarm systems integrated with a. voice sesistant 6 AUMLfortor: 1. The Internet of Things has been a fast-growing area in recent years. 2 Theuse of AVAL is 4 Al, machine learning and deep learning are being employed to tnake oT deve ad servicessmarter and more secure. 4. Inan industrial setting oT networks throughout a manufacturing lant can collect operational and performance data, 5, Itis then analyzed by AI systems to improve production syst improve production system performance, boost efficiency and predict when machines will require maintenance, 7. Aland Machine Learning in Hyperautomation : 1 "Hypertens at hing i an rdantaion thatcanbeastomaed sade suteesred means AWA sed mii eri 5 a ef rarer ig are key components and major drivers tehsil packaged software, 1ereasingly intertwined with IoT. Fautomation initiatives cannot rely on static ay 114M Sem-182) ‘An Overview to AL ‘Automated business processes must be able to adapt to changing circumstances and respond to unexpected situations. ‘That’s where AI, machine learning models and deep learning technology come in ‘Learning algorithms and models allows the system to automatically improve over time and respond to changing business processes ‘and requirements 8 Virtual agents: L Virtual agents have become valuable tools for instructional designers, computer application that interacts with 2 A virtual agent humans. 8, Web and mobile applications provide chatots as their customer service agents to interact with humans to answer their queries. 4, Google Assistant helps to organize meetings, and Alexia from Amazon helps to make your shopping easy. 5, Avirtual assistant also acts like a language assistant, which picks ‘cues from your choice and preference. 6, Virtual agents act as software-as-a-service too. | PART-5| 1 § Al and Ethical Concerns, Questions-Answers Long Answer Type and Medium Answer Type Questions Que 143,| What do you understand by ethics of artificial fence? | Taewer 1. The ethics of artificial intelligence is the branch of the ethi a i e is the branch of the ethics of tochnology specifi to artificially intelligent systems. Artificial Intelligence For Engineering 2. 1-15M (Sem-1 & 2) It is sometimes divided into a concern with the moral behavior of ‘humans as they design, make, use and treat artificially intelligent systems, and a concern with the behavior of machines, in machine ethics. Italso includes the issue of a possible singularity due to superintelligent AL ‘Que 1.14. | What are the ethical problems in Artificial Intelligence ? Following are some of the ethical problems that need to be dealt with in Artificial Intelligence A. Artificial Intelligence Bias : 1. Human beings are sometimes biased against other religions, genders, nationalities, ete. 2 This bias may unconsciously also enter into the Artificial Intelligence Systems that are developed by human beings. 8, The bias may also creep into the systems because of the flawed data that is generated by human beings. 4, For example, Amazon recently found out that their Machine Learning based recruiting algorithm wasbiased against women. 5, Itisimportant that AI researchers specifically try to remove bias while developing and training the AI systems and selecting the data, 6. There are many companies that are working towards cre unbiased Al systems, 7. IBMscientistshave created an independent bias rating system to calculate the fairness of an AI system. Rights of Robots : 1. If robots become advanced enough emotionally, should they be given equal rights like humans or lesser rights ? 2. What ifrobots kill someone. Should it be considered murder or & ‘machine malfunction ? All these are ethical questions that need to he answered as Artificial Intelligence becomes and more intelligent. Human Control on Artificial Intelligence : 1. Currently, human beings are the dominant species on Earth. 1-16M (Sem-1& 2) —V—_— ‘An Overview to AT 2, Human beings are dominant because of their intelligence. 3, So the critical question is, “What happens when Artificial Intelligence becomes more intelligent than Human Beings” 4. This would make AI the dominant species on Earth and lead to ‘huge changes in human existence or even human extinction. 5, Humans need to prepare for this scenario and how to deal with it. ‘Al replacing human workers : 1, As Artificial Intelligence becomes more and more advanced, it will obviously take over jobs that were once performed by humans, 2, Around 800 million jobs could be lost worldwide because of automation by 2030. 8, Now the question arises “What about the humans that are left unemployed because of this?” 4. Some people believe that many jobs will also be created because of ‘Artificial Intelligence and that may balance the sealesa bit. 5. People could move from physical and repetitive jobs to jobs that actually require ereative and strategic thinking. 6. But this is more likely to happen to people who are already educated and fall in the richer bracket. 7. ‘This might inerease the gap between the rich and poor even further. & Anew societal setup will have to be generated so that all human __ beings are able to earn money even in this scenario, Handling mistakes made by Al: 1. Artificial Intelligence may evolve into a super-intelligence in a few years but right now it is basic and so it makes mistakes. 2 Currently it make mistakes that are relatively safe and can be easily handled. 3, But Artificial Intelligence may make complicated mistakes in the future. 4, Then what is to be done ? The question is about relativity. Do Artificial Intelligence systems make lessor or more mistakes ‘than humans? 6. Do their mistakes lead to actually lose of life ? 1-17M (Sem-1& 2) Intelligence For Engineering 7. Andif there is a loss of life, is it more or less than when humans make mistakes ? | & All of these questions need to be taken into account when developing AI systems for different applications so that their mistakes are bearable and not catastrophic. Data & Algorithms @O0 CONTENTS History of Data. seu 268M to 2-2M. Data Storage and... 2-2M to 2-5M Importance of Data and its Acquisition ‘The Stages of Data vrenonninennsninanene 25M to 207M Processing Data Visualization... 27M to 2-9M. Regression, Prediction & ccrnwmon 2-9M to 2-12M Classification Clustering & Recommender ........ 2-12M to 2-15M Systems 2EM (Sem €2) Data& Algorithins PART: History of Data, Questions-Answers Long Answer Type and Medium Answer Type Questions Que2,_ | Explain data, information and knowledge. E=eal Data : Data are raw facts and figures that can be processed or stored by a computer. For example, text, numbers, symbols, ete Information : Information is data that has been processed into a form that tives it meaning. For example, analysis of retail of sale data ean provide information on which products are selling. Knowledge : Knowledges the understanding of rules needed to interpret information. For example, information on retail market sales can be analyzed ‘with promotional efforts to yield knowledge of customer behaviour. Applied for idan ee a purpose process Fig. 213 PART-2 Data Storage and Importance of Data and its Acquisition. Questions-Answers Long Answer Type and Medium Answer Type Questions Que 22, | What is data storage? “Answer L Data storage means that Det ans tht fs and documents are recorded digitally 294 ed in a storage system for future use. ried deal ‘Antficil Intelligence For Engin ng, 29M (Sem-1&2) Storage aystems may rely on eleetromagneti Jor other media to preserve and restore the data if needed. Data storage makes it easy to back up files for safekeeping and quick recovery in the event of an unexpected computing erash or eyber attack. Data storage ean occur on physical hard drives, disk drives, USB drives or virtually in the cloud. Que 2S. | Describe the types of data storage ? ‘Answer ‘Types of Data Storage : L Direct Attached Storage (DAS) : i Direct attached storage (DAS) includes types of data storage that ‘are physically connected to the computer. ii. Thisstorage is generally accessible to only a single machine, Some common devices in this category include 1. Hard Drives 2 CD/DVD Dri 3. Flash Drives fii, DAS solutions are great for creating local backups and can be ‘more affordable than NAS solutions, but sharing data between machines can be cumbersome Network Attached Storage (NAS): i. Network attached storage (NAS) allows for multiple machines to hare storage over anetwork. This is accomplished with multiple hhard drives or other storage devices in a RAID configuration. i. Thekey benefits of NASis the ability to centralize data and improve collaboration. fii, Data can be easily shared among connected machines, and permission levels can be set to control access. iv. While NAS solutions tend to be more costly than DAS solutions, they are still very affordable as storage technology has advanced significantly. Que 24. | Why is data important ? ‘Answer Importance of data: L ‘Data provides a deeper understanding of your market. 2. Data collection improves your consumer database. 24M (Sem-1 &2) Data Algorithms 3, Consumer data improves your marketing strategies. 4. Reallows for greater personalization. Que2.5) | What do you mean by Data Acquisition ? Teaver] Data acquisition isthe process of measuring physical world conditions aod phenomena such aa electricity, sound, temperature and pressure. o 2. ‘This is done through the use of various sensors which sample the eavironment's analog signals and transform them to digital signals ‘wing an analog-to-digital converter. ‘The resulting digital numeric values can then be directly manipulated by a computer, allowing for the analysis, storage and presentation of these data, [Que26. | Discuss the types of Data Acquisition Systems. [Answer ‘Types of Data Acquisition Systems 1. Analog Data Acquisition Systems : The data acquisition systoms ‘which can be operated with analog signals, are known as analog data acquisition systems. Following are the blocks of analog data acquisition systems: ‘Transducer: It convorts physical quantities into electrical signals. Signal conditioner It performs the functions like amplification and selection of desired portion ofthe signal. iti, Display device : It displays the input signals for monitoring purpose. iv. Graphie recording instruments : These can be used to make the record of input data permanently. ¥. Magnetic tape instrumentation : It is used for acquiring, storing and reproducing input data. 2 Digital Data Acquisition Systems : 4. The data acquisition systems, which can be of ith digit n stems, which can be operated with digital signals, are known as digital data acquisition systems. They Use igital eomponentsforstorngor displaying the information. The following operations take place in digital data acquisition : 4 Acquistion ofanalog signals Conversion ofanalg signals a analog signals into digital signals or digital data Processing of digital signals or digital data Artificial Intelligence For Engineering Antificial Intelligence ForEngincering __‘2 5M (Gem 2-5M(Sem-1 & 2) Following ortho Hacks of digital data aquisition ystems i. Transducer Tt onverts physical quantities ito elestrca signals. nal conditioner : It performs the functions lke ae sitcaon and selection of desired portion ofthe signa. si Multiplexer Connects one af th multiple input to output Staten pall ose converte. fv. Analog to Digtal Converter: converts th analog input toiteauvaen dial oat Display device: Ti diplasthe datain digital format, i: Digital Recorder: Is used to record the data in digital Pema PART-3. ‘ Pag The Stages of Data Processing. Questions-Answers oe Long Answer Type and Medium Answer Type Questions Que 27, What are the various stages of data processing ? ‘Answer ‘There are six main stages in the data processing cycle : 1. Data Collecti 4, The collection of raw data is the first step of the data processing eyele Tho type of raw data collected has a huge impact on the output produced. fii It is important that the raw data sources are trustworthy and well-built so the data collected is ofthe highest possible quality. iv, Rav data can include monetary figures, website cookies, proit/ loss statements of a company, user behavior, ete. 2 Data preparation i. Data preparation or data cleaning is the process of sorting and filtering the aw data to remove unnecessary and inaccurate data.. Raw datais checked for erorsand then transformed into a suitable form for further analysis and processing. 26M (Sem-1&2) Data & Algorithms zom(Semiea te iii Thisis done to ensure that only the highest quality data is fed into the processing unit. 3. Datainput : i. Inthisstep, the raw datais converted into machine readable form and fed into the processing unit. i Data inputs the first stage in which raw data begins to take the form of usable information. fi This can bein the form of data entry through a keyboard, scanner or any other input source. 4. Data Processing : i. During thisstage, the data inputted in the previous stage is actually processed for interpretation fi, ‘The raw data is subjected to various data processing methods ‘using machine learning and artificial intelligence algorithms to generate adesirable output. ii The process itself may vary slightly depending on the source of data being processed andits intended use. 5 Data output/interpretation : i. The data is finally transmitted and displayed to the user in a readable form like graphs, tables, vector files, audio, video, documents, etc. i. This output can be stored and further processed in the next data processing cycle. 6 Data storage : i. The final stage of data processing is storage. Afterall of the data is, processed, itis then stored for future use. i, This allows for quick access and retrieval of information whenever needed, and also allows it to be used as input in the next data processing yele directly. Que 28,_ | What are the different forms of data processing ? | Different forms of data processing are: 1. Data cleaning : Data cl leaning is a process to remove the noisy data, clean the data by filling in the missi ¥ Fratton bY Alling in the missing values and correct the 2 Dataintegration : Dat ion : Data integration is a technique that combines the data from multiple heterogenee ge feneous data i data store. Data integration ata sources into a coherent needs dataclesseg 7 ma Vol inconsistent data and therefore Artificial Intelligence For Engineering 2-7M (Sem-1 & 2) Artificial Intelligence For Engineering 0 OO 3% Data transformation :In this step, data is transformed or consolidated into forms appropriate for mining, by performing summary or aggregation operations. It involves the following. a. Smoothing : Smoothing is a process of removing noise from data, b. Aggregation : Aggregation is a process where summary or aggregation operations are applied to the data. c. Generalization : In generalization low-level data are replaced with high-level data by using concept hierarchies climbing. Normalization : Normalization scaled attribute data so as to fall within a small specified range, such as 0.0 to 1.0. Its of two types : i. Min-max normalization : It is a technique that helps to normalize data, It will scale the data between O and 1. i, z-score normalization : Transform the data by converting, the values to a common scale with an average of zero and a standard deviation of one. fe. Attribute/feature construction : New attributes constructed from the given ones. Data reduction : Data reduction is used to obtain reduced representation of data in small values by maintaining the integrity of original data. PART-4 Data Visualization. Questions-Answers Long Answer Type and Medium Answer Type Questions Que 29. | Describe data visualization. ‘Answer Data visualization is a set of data points and information that are represented graphically to make it easy and quick for user to understand, Data visualization is good if it has a clear meaning, purpose, and is very ceasy to interpret, without requiring context. 28M (Gem-1&2) 3 Que 210, | Write down the Characteristics of data Visualization. | ‘Tools of data visualization provide an accessible way to see and ‘understand trends, outliers, and patternsin databy using visual effects | fr elements such asa chart, graphs, and maps. Characteristics of data Visualization : 1 2 3 4 B It shows or visualizes data very clearly in an understandable manner, Teencourages viewers to compare different pieces of data. It closely integrates statistical and verbal descriptions of data set. It grabs our interest, focuses our mind, and keeps our eyes on message as human brain tends to focus on visual data more than written data, It aleo helps in identifying area that needs more attention and improvement. Using graphical representation, a story ean be told more efficiently Also, it requires Tess time to understand picture than it takes to understand textual data incuss the categories of Data Visualization. wer] Categories of Data Visualization : Data Visualization Y Categorical Data Visual ‘Nanerical Data Visualization { 1 { (Continaou Data Discrete ] [Binary ] [Nominal ] [Ordinary Data_||_Dsta Data & Algorithms | Artificial Intelligence For Engineering 29M (Sem-1 & 2) ‘Numerical Data : 1, Numerical datas also known as quantitative data. [Numerical datais any data where data generally epresentsamount such as height, weight, age ofa person, ete. ‘Numerical data vieualizationis easiest way to visualize data, Is generally used for helping others to digest large datasets and ‘aw numbers in away that makesit easier to interpret inte action. 5. ‘The type of visualization techniques that are used to represent, ‘numerical data visualization is Charts and Numerical Values. Examples are Pie Charts, Bar Charts, Averages, Scorecards, ete ‘Numerical datais eatogorized into two categories : a, Continuous Data : It can be narrowed or categorized (Bxample: Height measurements). 1b, Discrete Data:This typo of datais not “continuous” (Example: Number of cars a household has). ‘ategorical Data : ‘Categorical dat is also known as qualitative data. Categorical datais any data where data gonerally represents groups. Itsimply consists ofeategorical variables that are used to represent characteristics such as a person's ranking, a person's gonder, et. Categorical data visualization is all about depicting key themes, ‘establishing connections, and lending context. ‘The type of visualization techniques that are used to represent

You might also like