Weather and volatile market demand for crops are the two biggest concerns for farmers in developing countries like Nepal. Risk of crop production has worsened by changing weather patterns due to climate change. Conventional crop insurance systems are complex and often not feasible economically. Farmers are reluctant to get covered for their crops due to lack of trust in insurance firms and the fear of delayed or non-payment of claims. Farmers want rapid assistance when a natural disaster strikes so that plantation can be redone within the growing period. Crop insurance that is simple, reliable, and affordable is consequently critical, not only for farmer’s well-being, but also for the nation's food security. In this paper, a blockchain based platform for crowd farming with weather index-based crop insurance is presented. The solution presented is a platform that allows general customers to invest in agricultural projects in return for harvest from the field. To minimize risk from climatic disasters an affordable, efficient, low-cost crop insurance solution is also integrated which will ensure that farmers are insured and benefited from timely crop insurance. Currently, the expense of administering insurance is a considerable barrier to entry into this market. With the correct technology, this cost can be drastically lowered. Decentralized architecture and smart contract is the core foundation for achieving this platform since smart contract maintains trust among two involved parties and triggers unbiased payouts based on certain parameters. The success of this model can lead to developments in crowdfunding platforms, supply chain management system as well as other form of insurance

Education system has changed in many ways over the last decade, with the most important change coming in mode of learning and examination. Educational institutions are slowly moving into online teaching and education, which in a way, is changing the perception of learning among students and parents. Students often raise a query about which is better, online or offline exam. Online learning, e-learning, electronic teaching tools, and digital assessments are not innovations. However, there has been limited implementation of online invigilated examinations in many countries. The main aim of this project is to eliminate the logistical hassles and drawbacks in the traditional mode of the pen-and-paper examination. It reduces the time and the need to hire someone to check the students taking the exam. This project can be useful for giving/taking the exams, also during the pandemic without having to postpone the scheduled exam for any reasons. Being an integrated online examination system, it will reduce the paper work. It also assures the security during the examination using functionalities like facial recognition, tab switching etc. The AES Algorithm is used for encryption which will generate random numbers which shall be used as a passcode for accessing the exam paper. The js function will be used for the face recognition. The image uploaded during the sign-up time will be compared to the real time image of the student while appearing the examination and the face will be detected.

Semantic Parsing is a task of parsing through a structured data to get information. This task is important to get data from tabular structure content such as tables. The project focuses on using TAPAS architecture to get information from a table for a given English language query. The project is a sub task of Question Answering system and an extension to Text to-SQL system to obtain the user-desired answers from the content. In this project, a web platform is created which allows people to query a table-based data and get the answers from it. The application provides a way to upload a tabular sheet, modify the content, and then ask the question in plain English language. For the application, Google’s Table Parsing (TAPAS) architecture is finetuned on WikiSQL dataset. The system provides satisfactory output for the questions with the accuracy of approximately 84.76% on validation set. The system does operation such as search operation on string data and operations like sum, count and average on numerical data. The output also depends upon the choice of words used by the user. The system can take synonyms of some words which are contextual to a certain extent but struggles in non-contextual words. The system is also dependent on the names of the columns which are present in the querying sentence.

The world has experienced major growth in smart devices. Along with the growth in smart devices there has been significant advancement in AI and Machine learning. This has facilitated in developing various tools and technology to help each sector of human lives. The project is an attempt to help blind and visually impaired people to use their smart phones in performing basic tasks such as describing the environment, making emergency calls and even reading out the text. This attempts to make technology reachable to every individual. The rapidly growing development of mobile technology initially adversely affected the accessibility of device functionalities for visually disabled persons increasing their digital exclusion. Visually disabled persons through touch, gestures, voice commands and audio feedback can freely configure and run a set of applications available on a mobile device. The main goal of this project is to use the Machine learning algorithm to identify the environment. Another part of this system is to give the result automatically based on the recognized surrounding information. This paper describes the system that implements Machine learning and natural language processing to image captioning method, character recognition and carry some other useful functionality. As a result the system will assist visually impaired people to perform day to day activities.

Devanagari is an ancient script used for over 120 spoken Indo-Aryan languages, including Hindi, Nepali, Marathi, Maithili, Awadhi, Newari and Bhojpuri. This script is used by millions of people in India to write documents in Marathi and Hindi. Most of the Indian mythology is written in this script. Handwritten Devanagari character recognition has gained popularity over the years due to such importance of the script. Although significant research has been made in full character recognition of Devanagari characters using Convolution neural networks for both feature extraction and classification, the project experiments different classifiers for classifying and predicting the handwritten characters while using CNN and DNN for feature extraction. The scope of project has been widened by making the model to predict partial Devanagari characters while been trained on full characters and vice versa. Various algorithms like CNN, RNN and the techniques of image processing were studied to help complete this project.

Aim of the project is to present a platform where scenery images of real-world are transformed into cartoon style images in different styles. Sceneries specifically as they are widely used in various media like games/movies/anime/manga. This project will provide at least two styles for image cartoon-ization. The dataset used will be a selection of images with same artist / same style for accurate transformation of images into the particular artist’s style. The referenced paper belongs to learning based methods, which has been popular recently to stylize images in artistic forms. Amongst all the techniques usable, the application of a Generative Adversarial Network (GAN) is be used for the styling real_x0002_world images that uses 2 loss functions namely content loss and adversarial loss for getting a sharp and clear image.

The evolution of foreign ministries followed from the desire of rulers and their ministers to maintain a continuous flow of diplomatic business in which cross relationships between diplomatic partners, between internal sources of political influence and between differing issues could be carefully followed and controlled. International relations is by definition the study of countries' interactions with other countries. Since at least the time of the ancient period, these relationships have been directed, promoted, and negotiated by way of diplomacy and diplomats. Unsurprisingly then, scholars have devoted considerable attention to the subject of envoys and their activities. As with all research efforts, data is key to progress. However data on diplomacy, whether qualitative or quantitative, can be difficult to obtain and difficult to work with. The reasons for this paucity are well rehearsed. For one thing, diplomatic records are often covered by countries' secrecy laws, and may be made available for public consumption on a haphazard schedule, if at all. Simultaneously though, governments often release files in overwhelming numbers but without careful curation, making it impossible for scholars to keep up with the flow of information available to them. On the one hand, the sheer size of the data collection makes it difficult to catalog and work with, especially if one wishes to move between aggregate analysis and the inspection of individual cases. On the other hand, finding more than one example of a particular phenomenon and applying the scientific method to it can be daunting. All told, there is an obvious need for a classification of Diplomatic Missions Data using Machine Learning. The TF-IDF(Term Frequency-Inverse Document Frequency) algorithm that has been used for the project easily helps to classify the diplomat data into the necessary categories.

As we move into a world dominated by technology, data as well as processes are digitized to create a more efficient system and simpler life. The information we required can be found within some clicks. Taking an example, contact information directory are no longer physically required. A single smart device like smart phone with access of internet can handle this. Similarly, it is not always possible for one to carry physical directory while in day to day life and needed information may be found very slowly. In order to gain the related information, one always gets deprived of fast information deliverance which this project aims. A web application for general user and mobile application for admin / listers is made for implementation of the system of online business directory. Autocomplete search feature will be used to search required data in order to gain desired result in a quick time. Autocomplete is a search feature where the search engine predicts the user's query and provides suggestions as the user provides starting input. Predictions and suggestions are generated on the basis of search history by overall user and rank of data. The users query are predicted by the system on the basis of search history.

A Machine Learning Model is a file that has been trained to recognize certain types of patterns. You train a model over a set of data, providing it an algorithm that it can use to reason over and learn from those data. Once you have trained the model, you can use it to reason over data that it hasn't seen before, and make predictions about those data. Machine Learning is a branch of Artificial Intelligence giving computers the ability to model various types of data using different types of algorithms. Among its various application areas, use of machine learning to perform real-world object detection has been explored heavily for PC as well as smartphones. As we move into a world dominated by technology, data as well as processes are digitized to create a more efficient system and simpler life. Map, phonebook, camera, music player is no longer physically required. A single smartphone has the functionality to perform all the tasks that they do. Similarly, it is not always possible for one to carry huge and heavy books while travelling in order to gain the related information, one always gets deprived of fast information deliverance. There are various people who love travelling and exploring the nature might encounter various species of animals. Animal detector also replaces the troublesome and traditional way of learning as well as research on wildlife especially for the people related to the field of Zoology. One can get speedy information just at a matter of few taps on their mobile screen using this application.

A Machine Learning Model is a file that has been trained to recognize certain types of patterns. You train a model over a set of data, providing it an algorithm that it can use to reason over and learn from those data. Once you have trained the model, you can use it to reason over data that it hasn't seen before, and make predictions about those data. Machine Learning is a branch of Artificial Intelligence giving computers the ability to model various types of data using different types of algorithms. Among its various application areas, use of machine learning to perform real-world object detection has been explored heavily for PC as well as smartphones. The face is one of the easiest ways to distinguish the individual identity of each other. Face recognition is a personal identification system that uses personal characteristics of a person to identify the person's identity. Human face recognition procedure basically consists of two phases, namely face detection, where this process takes place very rapidly in humans, except under conditions where the object is located at a short distance away, the next is the introduction, which recognize a face as individuals. Stage is then replicated and developed as a model for facial image recognition (face recognition) is one of the much-studied biometrics technology and developed by experts. There are various kinds of methods that are currently popular in developed face recognition pattern namely, Eigenface method, LBPH , Fisherface method. Facial image recognition LBPH method is based on the local_x0002_binary-operator, broadly implemented in face recognition, due to its discriminating strength and calculation easiness for facial features. Haar Cascade Classifier uses LBPH algorithm for the Extraction of feature of human face. Face recognition library recognize and manipulate faces from Python or from the command line.

This paper discusses the implementation of an online voting system based on image steganography and visual cryptography. The concern of the trust deficit of people in online voting is addressed by providing security and confidentiality through Steganography and Visual Cryptography. The system is implemented in Django Framework on a web-based interface, with an SQLite database server. After considering the requirements of an online voting system, current technologies on electronic voting schemes in published literature were examined. Next, the cryptographic and steganography techniques best suited for the requirements of the voting system were chosen, and the software was implemented. The system is incorporated with techniques like the password hashed-based scheme, visual cryptography, image steganography, and threshold decryption cryptosystem. The analysis, design, and implementation phase of the software development of the voting system is discussed in detail.

Blockchain technology is a revolutionary innovation for its potential to build solutions where outsiders can transact with each other without being dependent on any middle person to supervise the transaction between the parties. Blockchain solutions attempt to solve the crucial matter of user privacy, albeit that blockchain was initially directed towards full transparency. In the context of Know Your Customer (KYC) standardization, a decentralized schema that enables user privacy protection on blockchain is proposed with Public smart contracts. Through the public KYC smart contract, a user registers and uploads their KYC information to the exploited IPFS storage, actions interpreted in blockchain transactions on the blockchain. The presented system introduces effectiveness and time efficiency of operations through its schema simplicity and smart integration of the different technology modules and components. This developed scheme focuses on blockchain technology as the most important and critical part of the architecture and tends to accomplish an optimal schema clarity.

Technology is evolving in rapid speed. Machines has started to learn on its own. The use of Machine Learning in different field has been increasing day by day. Image-to-image translation is a class of vision and graphics problems where the goal is to learn the mapping between an input image and an output image. It can be applied to a wide range of applications, such as season transfer, style transfer, and photo enhancement. This project shows how an algorithm can be implemented for image-to-image translation. This project helps artist to use their mobile devices to draw a sketch and help them to visualize how their art (Sketch) looks like in real world i.e., to convert imagination to reality. The main goal of the project is to use machine learning to study the image and convert the image form one domain (sketch) to another domain (real look alike human face). In this project, the model has been trained under various condition given different value of generator loss to choose the model for the application. In this project Pix2pix algorithm has been used for image-to-image translation. PSNR value is taken into consideration for quantitative analysis of generated image

Communication is essential for achieving managerial and organizational effectiveness, employees will not be able to aware of what their co-workers are doing, will not have any idea about what their goal are, and will not be able to assess their performance. In absence of channels of communication, supervisors will not be able to give instruction to their subordinates and management will not receive the information it requires to develop plans and take decision. Chatting is a method of using the technology to bring people and ideas together despite of the geographical barriers. The technology has been available for years but the acceptance it was quite recent. Our project is an example of an android chat. It is made up of application, the client application which runs on the user mobile. To start the chatting our client should get connect to the server, where they can do group and private chatting the required software and hardware are easily available and easy to work with. Basically, the project describes how to manage for good performance and better services for the clients. Flutter was used for the frontend of the project and for the backend I have used firebase which both are provided by Google. This made me finish my project with ease. Socket based approach is used for the message transition between two users. Images can be picked from the gallery as image picker is used which can be used to share images between users. From this above I created a well working chat application for friends, and family to communicate with each other and have a good communicable vibe between them. At the end of this project, I was able to manhandle a single project which was a huge confidence booster for me.

NULL

Breast cancer, cervical cancer and PCOS (polycystic ovary syndrome) are the most common reproductive health issues found in women [1]. In an underdeveloped country like ours, women are under-privileged by the dominance of men. However, due to orthodox beliefs and lack of awareness women restrain themselves from going to the hospital for a proper checkup and discover their conditions only at the advanced stage. The main objective of this project is to bring all the common health problems of women into one web app and try to change the modality of health checkup and increase its awareness. This project helps to detect the presence of metastatic tissue and invasive ductal carcinoma in the breast by the help of histopathological images. Histopathological images are the images under the microscopic examination of a biopsy or surgical specimen that is processed and fixed onto glass slides. Traditional method of detecting breast cancer involves manual observation of pigmented tissues under microscope, which is a tedious and time-consuming task. It might also not always give accurate prediction. Hence, Convolutional Neural Network is used to extract features and then classify using a fully connected network. It also helps in detection of cervical cancer and PCOS in women by performing preliminary tests using Ensemble Classifier model, which combine the decisions from multiple models like decision tree, support vector machine (SVM), random forest, and logistic regression models to improve the overall performance. As a result, this app is useful for hospitals for the quick prediction of breast cancer, cervical cancer and PCOS with a maximum accuracy. This app has a user friendly UI design so it is useful for any one to predict if she has cervical cancer and PCOS. With this, women are able to detect it in an early stage and prevent further complications.

A role-playing-game is a game in which players assume the roles of characters in a fictional setting. The fantasy war games in the 1970s gave a rise to the modern role-playing-game. This genre of the game popularly provides a gist of action and a sense of adventure and accomplishment to a player. Therefore, this project is carried out to develop a 3D game that has an open-world level design to provide an immersive experience for the players. In addition to that, the gameplay is very simple and easy to get used to. The game is designed using Unity game engine. Unity is a synthesizing type of game engine that the designers could use to develop a videogame, visualized constructions and real-time 3D animations. Furthermore, it is a cross-platform engine that supports multiple operating systems like Windows, Linux, Mac, iOS, Android, etc. It provides an interface for controlling the game assets with scripting. It provides a standard set of libraries for game development and allows scripting in C# language. The project aims to develop a 3D action combat game from the top down perspective that focuses on implementing rigid body physics, movement mechanics, lighting, etc. The project demonstrates the basic features of the Role Playing Game genre of games and the process of 3D game development.

As per the research, reading the bad news does not have good impact on reader’s mind. This has introduced the need of creating an application that can let a reader know if the news is good or bad. There are a lots of news portals where news is being published every day and there is also a substantial competition between the news publishing site. In order to have more audiences, the news sites publish everything without looking for the sentiment of news. People perception tends to change as per the news content they read. There are lot of negative news surfacing around the news portal. Continuous consumption of negative news can lead to anxiety and lots of negativity. The number of positive news have been outclassed by negative news. In this project machine learning has been used for news classification purpose. This machine learning approach classifies a news in to three different categories i.e., positive, negative and neutral by analyzing its headline. An authentic site has been chosen in order to implement these classifications.

Recommender systems are software applications that provide or suggest items to intended users. The content-based technique is adopted because of its suitability in domains or situations where items are more than the users. Content-based recommenders provide recommendations by comparing the representation of contents describing an item or a product to the representation of the content describing the interest of the user (User's profile of interest). They are sometimes referred to as content-based filtering. The content-based technique is adopted or considered here for the design of the recommender system for project documentations of different CSIT Colleges. The content-based technique is suitable in situations or domains where items are more than users. There are many projects conducted in various CSIT Colleges in Nepal. These projects are conducted on a yearly or semester basis. This application acts as a repository or an archive to store those projects where students, as well as different academic professionals, can view them. This android application provides documentation to the user with the help of Content_x0002_based Filtering. It helps the user to view large amounts of data as the application will display it according to the user’s preferred or selected subjects. It helps to narrow down the domain for each user. With this application valuable information can be stored for a long period with easy access.

Traditional relational database management solutions (e.g. Oracle and SQL), deployed globally across millions of applications, have one major operational constraint – the management of data is performed by a few entities who must be trusted. Distributed Ledger Technologies (DLT, commonly referred to as blockchain), an alternative architectural approach to managing data, and removes the need for a trusted authority to store and share a perpetually growing set of data. A foundation characteristic of a blockchain is trust. Blockchains have digital signatures and use keys to authorize and check transactions and positively identify the initiator. Cryptography hashing like SHA256 are used to securely hash the blocks such that they cannot be tampered. And the consensus protocol makes sure that all the nodes in the chain work on the same chain of data. As in organization or(government), transparency in the case of the fund allocation is very low and we don’t have good system to check for the allocation and tracking of the fund System can be upgraded for the case of fund allocation and tracking with the help of blockchain technology. This technology is not just limited to the fund allocation we can easily derive it to use for supply chain .Blockchain principle is modified according to the organizational need for the required system to perform in given organization allocation and funding system. Login based system is created along blockchain principle for transparent allocation. A decentralized fund allocation and tracking system over P2P network with the help of blockchain running smart contract is presented so that the strong highly transparent system can be established in the case of fund allocation and tracking.

In the past decades, the world has gone through significant financial change. With the growth of Bitcoin and other crypto currencies, people are shifting their interest toward crypto currency. The way of raising the funds for the projects has also gone through the change. Instead of IPO, people are more interested in ICO for funding the project using cryptocurrency. This paper focuses on creating my own ERC-20 token using the open Zeppelin library in Ethereum Blockchain. For the created token, ICO platform is created for raising the funds of project. The entire project is done on the rinkeby test network which means no real crypto currency is used in this project. This project is a prototype for digitizing the assets into crypto tokens and raising the funds using those tokens. The total supply of fixed 10 million token (Dyawa) has been used in this project. This developed prototype can be used by any startup company who wants to digitize the asset into crypto tokens and raise the funds for starting their projects. For implementation in real life, the supply of token and policy of token of developed prototype needs to be changed as per the business requirement. Furthermore, rules and regulation of governing body like SEC needs to be followed strictly while implementing in real world scenario.

“Online Examination And Evaluation system” is an web based application built for conducting entrance and written examinations through the online medium using laptop or desktop. The purpose of Online Examination And Evaluation System is to automate the existing manual system by the help of computerized equipments and full-fledged computer system, so that student should not visit physical location to give the exams and the exam data can be stored for a long period of time with easy accessing from anywhere. “Online Examination And Evaluation System” as described above, can lead to error free, secure, reliable and fast management system. It can assist the user to concentrate on their other activities rather to concentrate on the record keeping, monitoring the exam and so on. Thus it will help organization in better utilization of resources. Examinee can be benefited as they no longer need to visit organization for exam and result and can use that time in other activities. This web application is build using the react framework of the javascript for the frontend and django rest framework of the python programming language as backend. This web application require the safe exam browser to take the exam. Safe exam browser is an open source web browser. In this web application the student can register for the entrance exam and all other student and teacher registration is done by the admin. The teacher can add the exam question and can check the student answer after student has taken the exam and mark the student answers. The student can take the exams and see the marking given by the teacher and the admin can manage the overall system.

There are various reasons due to which the world is changing. This is causing various natural calamities in various parts of the world. The scale and timing of these calamities are like never seen before. For example: Wildfire in various parts of Southern Nepal, Wildfire in Amazon, and Typhoon in Philippines etc. Therefore a real time disaster tracker seems to be necessary for the alert and information regarding these calamities. Thus “Natural Disaster Tracker” is made to meet this requirement. An application that shows weather forecast and natural calamities. Thus “Natural Disaster Tracker” is made to meet the demand. An application that shows weather forecast and natural calamities. All in all this application can be used by government organizations such as the Meteorology division as well as general public for the NRT information. NRT stands for Near Real Time. As one must be quick to act in such conditions, timely information of these disasters are a must. Near Real Time information is provided in the application with the means of manmade satellites orbiting the earth. The information is extracted through satellite imagery which takes place in outer space. Due to this process a slight delay in the processing of information happens which leads to the information being Near Real Time. Such information can be used and implemented by authorities and personals to evacuate the respective areas and protect the life and property. This can result into a great tangible difference Information Technology can bring to human life.

Electronic Mail is one of the most convenient and reliable method for communication and inexpensive way for communication regardless of the distance. However, increasing volume of unsolicited emails is degrading the productivity. Electronic spamming is the use electronic messaging systems to send an unsolicited message (spam), especially advertisements, as well as sending messages repeatedly on the same website. While the most widely recognized form of spam is email spam, the term is applied to similar abuses in other media: instant messaging spam, Web search engine spam, spam in blogs, wiki spam, mobile phone messaging spam, junk fax transmissions, social spam, spam mobile apps, etc. The Naïve Bayesian Spam Classification has been used in this project where it designed to detect spam emails and separate them from ham emails. A Bayesian network is representation of probabilistic relationships. The algorithm was trained using Enron Dataset which is a very known spam email dataset. This project will show that Bayesian filtering can be simply implemented for a reasonably accurate text classifier and that can be modified to make a significant impact on the accuracy of the filter. A web application is developed which would input the emails by the user and receives the predicted probability that if the given email is spam or ham. The output obtained will provide the prediction if the email is either spam or ham as per the datasets. Experimental results have been collected using Enron dataset that consists of total 33,316 data including both spam and ham. The accuracy obtained using Naïve Bayes classifier is Bayes classifier is 98.81%.

In this era of Artificial Intelligence, humans are becoming more dependent on technology. With the enhanced technology, multinational companies like Google, Tesla, Uber, Ford, Audi, Toyota, Mercedes-Benz, and many more are working on automating vehicles. They are trying to make more accurate autonomous or driverless vehicles. Information on self-driving cars is everywhere, where the vehicle itself behaves like a driver and does not need any human guidance to run on the road. This compels us to think about the safety aspects—a chance of significant accidents from machines. Many researchers are running on various algorithms to ensure 100% road safety and accuracy. One such algorithm is Traffic Sign Recognition that is proposed via this project. When we are on the road, we see various traffic signs like traffic signals, turn left or right, speed limits, no passing of heavy vehicles, no entry, children crossing, etc., that we need to follow for a safe drive. Likewise, autonomous vehicles also have to interpret these signs and make decisions to achieve accuracy. The methodology of recognizing which class a traffic sign belongs to is called Traffic signs classification. In this project, a model has been built for the classification of traffic signs available as images into many categories using a CNN and Keras library.

“Hamro Barnamala” is an Mobile Based Application developed to provide a user friendly and entertaining digital platform for the kids that teaches Nepali language from basic to advanced. The Mobile application consists of huge amount of data’s related to Nepali language like Byanjans, Barakhari, Swor, Sankhya where every section of Nepali language contains images, audio and video which helps kids to learn more easily and quickely. The main aim of “Hamro Barnamala” is to teach kids those who are deprived of getting education especially from rural areas free education about Nepali language. As every data will be available offline in mobile application even if children don’t have access to mobile phones or over internet only one mobile is enough for those children to get education which can be of their parents or neighbors. There are two different views in “Hamro Barnamala”: Admin view and the Client or Mobile view. The Admin view is meant for the administrator to add, update, remove data’s like images, audio, videos and also manage users. The User will be able to fetch data’s as they require from mobile application, play quizzes and games which makes learning more fun. Also, the mobile view will be accessible to the clients or users, and they will be able to handle their information such as their name, address, and contact.

E-commerce (Electronic Commerce) is a vital platform nowadays for online business. In this era of internet, e-commerce is growing at a high rate keeping the growth of online business. For increasing the use of e-commerce in developing countries the B2B e_x0002_commerce is implemented for improving access to global markets for firms in developing countries. The primary goal of ecommerce is to reach maximum customers at the tight time to increase sales and profitability of the business. This paper outlines multitenancy based ecommerce platform similar to shopify where subscription-based packages allows anyone to set up an online store based on sudomain and sell their products using the platform. The entire development is primarily divided into two parts - the front-end development and the back-end development. The database design is also discussed with an emphasis on its relational connectivity.

The daily use of the internet for e-commerce has exploded in recent years and will continue to do so. Supply Chain Management is a network of facilities that create raw materials, transform them into intermediate items, and then deliver final products to clients via a distribution system. The current condition of online product buying and selling, customers have no way of knowing whether the things they're buying are from the same manufacturer. To deceive consumers into buying a product, many product resellers appear as legitimate manufacturers. The project demonstrates how blockchain and cloud services can be used to ensure that the goods a customer is purchasing is legitimate and comes from the original manufacturer. The project uses advanced technological stack containing AWS cloud service which use serverless architecture. The main goal of this project is to create a secure way of handling the supply chain from manufacturer to the end customers, ensure that the integrity of the product is maintained/protected.

A role-playing game (abbreviated RPG) is a game in which players assume the roles of characters in a fictional setting. This type of genre of game has been mostly based on the fantasy war games which mostly filled with violent elements such as fights. The rise of non-violent role playing games have led to creation of many games in role playing genre that focus on the emotional, easy-going aspects of the game. Therefore the project aims to create farming role-playing game which is a 2D (Two-Dimensional) farming game that simulates farming in a two-dimensional environment. In recent years farming games have become a common form of entertainment and way to relax as it helps to simulate farmers' lifestyle somewhat. The game is designed and developed using Unity game engine. Unity is a type of game engine that the designers could use to develop a videogame, visualized constructions, and 2-D simulations. Furthermore, it is a cross-platform engine that supports multiple operating systems like Windows, Linux, Mac, iOS, Android, Switch, XBOX, and PlayStation etc. The game engine provides an interface for controlling the game- objects (asset, image, physics) with scripting. The engine comes with libraries for game development and allows use of C# as its main scripting language. This report aims to show the implementation of 2-D Farming RPG. The implementation, methodology, and feasibility analysis are discussed in this report.

Automated parking system is a system which is created with the motive of providing better parking facilities to the riders. The application lets the user view the parking locations and the available parking spots. This helps users choose their own parking spot. Whenever a vehicle enters the parking station, the number plate of the vehicle is detected by the help of cameras. The vehicle number is then stored in the database with the arrival time. During the exit of the vehicle the plate is again detected and the exit time is stored in the database. This is used to calculate the total parking cost and is paid either in cash or through online payment methods. I hope that this system will contribute towards the betterment of the parking experience.

Online voting is a new trend that is gaining momentum in modern society. Online voting system has a great potential to decrease organizational costs and increase voter turnout. One main advantage over the traditional voting system is that it drastically decreases the printing cost of ballot paper or opening polling stations. Voters can vote from anywhere if there is an Internet connection. Despite carrying a great potential, online voting system are viewed with a great deal of caution because they introduce several threats. A single vulnerability on a centralized system can lead to large scale manipulations. Blockchain based voting system is a replacement for traditional electronic voting system. Using Blockchain based voting system we can introduce non-repudiation, distributed and security protection characteristics as it offers decentralized nodes for electronic voting. Blockchain is one of the emerging and challenging technologies with strong cryptographic foundation enabling application to leverage resilient security standards and distributed database. With the help of distributed approach provided by the Blockchain, an online application that can be used for Voting is created. The application provides an easy interface for Voters and Organizing Committee to organize an Online Voting.

The project is an attempt to help provide network and resources through a podcast directory platform to promote highly enthusiastic and growing creators, rich with contents to distribute among Nepalese as well as globally. The project constitutes of various features which allows efficient surfing over the listed podcast. It allows people to listen to podcast and manage them according to subscribes, playlist, history and favorites. The project allows admin to create podcast through RSS (Really Simple Syndication) feed. The system extracts the XML, parses it and then stores the information into the database. This allows admin to insert newer podcast without hassle. The update process also uses hashing algorithm to check for newer updates and updates the whole podcast by checking the hash values. This allows admin for easy update of existing podcast. The system constitutes of recommendation engine which gives recommendation based on content based recommendation. The recommendation is given comparing the podcast’s category, genre, language and description. Also the recommendation is provided for the episodes of a podcast. The project also has a live audio streaming platform which is based on real time audio broadcasting.

Computer vision is a branch of Artificial Intelligence giving computers the ability to view the real world and make sense out of it. Among its various application areas, use of computer vision to perform precise measurement of real-world object has been explored heavily for PC as well as smartphones. For smartphones, AR technology is predominantly used for developing applications with such functionalities. However, little research has been done to explore how measurement of objects can be performed precisely in absence of devices that support AR technology. This Project was developed upon realizing that ArUco could be used to perform the same functionalities for smartphones without AR support or depth sensing cameras. Utilization of different image manipulation functions in Opencv-Contrib and other python libraries to develop a Flask API was done in order to detect object in a plain background with an ArUco marker as a reference object. The User Interface (UI) of the project was implemented using Flutter Framework. Algorithms such as the Canny Edge Detection, Circle Hough Transformation and Angle Detection using Three points from 2D image were successfully implemented in backend portion of this project. After successful implementation of such algorithms and technologies, measurements could be performed. The inability to detect objects due to presence of shadows or low contrast between object and the background was a drawback of the project. The project can serve as an alternative to physical tool for length and area estimation to individuals in mathematics or engineering background.

Chess is an ancient game initially name Chaturanga originating to sixth century in India. Chess has come a long way from a normal two player game to a competitive environment. Playing chess helps to build our strategic thinking and levels up our skill of analysis. The motive of every chess player is to increase their strategic thinking and to beat the opponent. In order to do so, the player needs to study (analyze) their game and game of top players. This project implements the Alpha-Beta Pruning algorithm in order to analyze the game and give out the best move in any certain situations. This helps the player to know the best move and moving forward he is able to understand the why the move is best. The Alpha-Beta Algorithm backtracks to certain specified depth and finds out the possible best move in any situation. The implementation, methodology, and feasibility analysis are discussed in this report.

Pomodoro technique is a great time management technique. However, rather than the basic timer functionalities most Pomodoro Systems don’t add enough features to develop it as a full-fledged system. As a PWA optimized for offline and cross-platform functionalities using the caching locally through progressive web capabilities, it can form a niche system. Such a system would not need a constant internet connection to perform the basic of functions. It can still rely on the network when available but fall back on the cache for all its different functionalities. This system can act as a hub for the pomodoro technique on multiple devices such as desktops, tablets and smartphones. Vue JS forms the backbone for the core components of the application. Vuex helps to maintain state across all components in a consistent way. Offline capabilities are achieved through caching mimicking installations like that of current android applications.

Electronic Commerce is process of doing business through computer networks. A person sitting on his chair in front of a computer can access all the facilities of the Internet to buy or sell the products. Unlike traditional commerce that is carried out physically with effort of a person to go & get products, ecommerce has made it easier for human to reduce physical work and to save time. E-commerce which was started in early 1990’s has taken a great leap in the world of computers, but the fact that has hindered the growth of e_x0002_commerce is security. Security is the challenge facing e-commerce today & there is still a lot of advancement made in the field of security. Recently the e-commerce platform is playing an important role in some areas; its activities are a subset of e-business activities. The main advantage of e-commerce over traditional commerce is the user can browse online shops, compare prices and order merchandise sitting at home on their PC. For a developing country advancement in the field of e-commerce is essential. The aim of this paper is to build and develop a reliable website based on the e-commerce theories, developing effective well designed web pages. This website will sell All products include (like Computers, Laptops, Tablets). For implement the selling online website, it needs to use current technologies to achieve this goal. As a first stage, it should set up online ecommerce store with easy-to-use. Then improve the customer experience, and lastly implement the Direct Online Sale between business to consumer by implement electronic payment methods. All these techniques should be based on deliberated plan according to strategy of electronic ecommerce with implement the current technology to ensure a good revenue to the company.

Visualization aids better understanding and creates a unified and comprehensive tool for learning. Thus, the role that visualization plays in an individuals' mathematical thinking and problem-solving experiences has become more significant. Visualization of algorithms helps to assist learners to have a deeper understanding of different algorithms, compare their usage and apply them in various applications like maps, routing packets over the internet and game development. Visualizer for Pathfinding and Sorting Algorithms can be used as an educational tool that allows learners to visually inspect the behavior of various sorting and pathfinding algorithms based on graph theory. The main aim of this system is to demonstrate the workings of different pathfinding algorithms to find the shortest path between two points on a map based on graph theory and to demonstrate the workings of sorting algorithms. After selecting the algorithm to be visualized, the user can run an automated animation to visualize the functioning of the algorithm. The pathfinding algorithms included here are Breadth First Search, Depth First Search, A* Search and Bidirectional Search. The sorting algorithms included are Bubble Sort, Selection Sort, Insertion Sort and Merge Sort.

Education system has changed in many ways over the last decade, with the most important change coming in mode of learning and examination. Educational institutions are slowly moving into online teaching and education, which in a way, is changing the perception of learning among students and parents. Students often raise a query about which is better, online or offline exam. Online learning, e-learning, electronic teaching tools, and digital assessments are not innovations. However, there has been limited implementation of online invigilated examinations in many countries. This paper describes a system which implements blockchain along with could communication platform as a service, to monitor the authenticity of the document uploaded by the students at a specific period of date and time. The system will use the AES symmetric block cipher and SHA-256 hashing algorithm to safeguard the question papers and answer sheets respectively. As a result, the system will also ensure organizing the exams offline with the help of cloud service used as SMS. Furthermore, it will also not use the system time to track the submission of the system as it can be changed.

This project shows how we can implement an algorithm for the detection of handwritten expression evaluators to build a system that will evaluate the results of the raw user input or image. As we know, we have to manually calculate handwritten expressions which are tedious and time-consuming. In the traditional method, the user has to use a calculator or calculate the result of the expression themselves. Not only that student from primary levels is learning the mathematical expression using traditional memorizing methods which is not as efficient as the learning done with more exciting and entertaining way. Therefore, the proposed solution is to develop a working prototype of a system that provides result based on the handwritten digits and expressions with its value. The main goal of this project is to use the Convolutional Neural Network (CNN) algorithm to identify the handwritten number. Another part of this system is to give the result automatically based on the recognized digit and expression.

This study investigates the learning effects of playing racing, action, and sports computer games. In particular we focus on traffic school students’ driving behavior. A survey conducted at three driving schools, questioned driving students about their gaming habits. The driving instructors evaluated their students’ driving skills and traffic safety attitudes. The results indicate that experience in computer games can have a positive effect on driving performance. Experienced gamers were ranked significantly higher by their instructors regarding their overall driving skills compared to students with low experience in computer games. However, no evidence was found to indicate that experienced gamers have a worse attitude towards fellow road-users or traffic safety. Experiments conducted in a driving simulator, using a game developed purposely to enhance certain traffic safety variables, reveals that it is possible to provide an entertaining game with serious content. Preliminary results, however, indicate that the version of the game where the explicit game goals are hidden was found to be the most entertaining one. The results of the investigation warrant further review into the development and utilization of computer games for traffic safety and education purposes