Here in this post we will give you some CSE Mini project Ideas that you can definitely try to implement in this year .Some times being a student we face difficulty in choosing the project.So here we cover CSE mini project topics ,problem statement of some project .You can even check the papers of this related Project ideas and get some ideas to design your new algorithm or gain some new concept to it.We also added paper links for your referrence.
Hopefully this list of CSE mini project ideas will help you select a domain of your next project.Here below we give you some project ideas on Machine Learning,Deep learning and Computer Networking.So choose your poject wisely.:-)
CSE Mini project Ideas
- CSE Project Topic/Ideas 1 : Deep Network for Scene Understanding
Semantic segmentation is enough for object detection by defining each pixel and create boundary boxes to identify object, but we can efficiently detect and label objects using pipeline in which the result generated by semantic segmentation will be input to the object detection module for scene understanding. In semantic segmentation, object is identified by labelling each pixel and localize and detect object using ground-truth segmentation whereas in object detection, objects are classified by creating boundary boxes. This project will implement both semantic segmentation and object detection via pipeline system with the help of conventional neural network. There are many predefined neural networks for object detection, but implementing both object detection and semantic segmentation will gives efficient result in less time and in less cost.
- CSE Project Topic/Ideas 2 : Evaluation of BLUE queue management algorithm using ns-3
The main idea of this project is to evaluate the effectiveness of BLUE queue management algorithm in controlling the queue delay by using ns-3. Queueing delays are increasing resulting a situation called Bufferbloat. Bufferbloat occurs when a network link becomes congested, causing the buffering of packets. Bufferbloat reduces the overall network throughput. The network congestion causes extra delay apparently in the modern network equipment. Due to the fall in memory prices, the size of buffer in the equipment has increased.
The TCP congestion avoidance algorithm works on the packet drop mechanism to determine the bandwidth available. A TCP sender increases the rate at which it sends packets until packets start to drop, then decreases the rate. For the sender to select a suitable rate, the packet must be dropped timely. But due to the reduction in the price of memory and leading to large buffer capacity of the routers, the packets get queued for a long time. No packets are dropped resulting an increase in queueing delays. So, there is a need for active queue management algorithm which can control the queue delay within limits.
- CSE Project Topic/Ideas 3 : Elliptic Curve Cryptography for Secured Text Encryption
In the digital world the we have created for ourselves, our online lives have become more entwined with our social lives, it’s more critical than ever that our professional and social messages are secured, particularly if we’re talking
about sensitive information. Because of the comfort aspect, more businesses are starting to use messaging apps and group chat features as proxies for in-person conversations. As a result, it’s critical that our preferred networks are protected from hacking or snooping, particularly as our economy becomes increasingly dependent on information and data. Encryption is a good way to keep messages secure when they’re being sent over the Internet. Information could be extracted as it transmits between two points, before widespread adoption, making our data vulnerable to hacking or surveillance.The data we send over the internet is, therefore, very prone to getting sabotaged by attackers. This gives rise to the demand of a secure encryption technique to securely send text over the network without it being sabotaged by predators.
The problem of securing transfer of data on the internet is a critical one in present world where everything is digital.
The problem statement; to find an encryption technique for text to ensure only the intended people can have access to it. Elliptic curve cryptography is used to solve this problem.
- CSE Project Topic/Ideas 4 : Automated Detection and Classification of Apple Leaves Diseases
Scab and Rust are the two common types of apple leaf diseases. Early diagnosis and accurate identification of apple leaf diseases can control the spread of infection and ensure the healthy development of the apple industry. The existing research uses complex image preprocessing and cannot guarantee high recognition rates for apple leaf diseases. This project we used an accurate identifying approach for apple leaf diseases based on deep convolutional neural networks.
It includes generating sufficient pathological images and designing a novel architecture of a deep convolutional neural network to detect apple leaf diseases. Using a dataset of 1821 images of diseased apple leaves, the proposed deep convolutional neural network model is trained to identify the four common apple leaf classes. Under the hold-out test set, the experimental results show that the proposed disease identification approach based on the convolutional neural network achieves an overall accuracy of 91.71%. This project indicates that the proposed deep learning model provides a better solution in disease control for apple leaf diseases with high accuracy and a faster convergence rate.
- CSE Project Topic/Ideas 5 : QoS Enhancement using machine learning based route selection in MANET
MANETs (mobile ad hoc network) consists of highly dynamic and mobile set of nodes. With its dynamic topology and variable parameters, the MANETs lack QoS support. Various real-world applications said above requires some degree of confidence in end to end packet delivery. And some applications need time constrained delivery in interactive applications.
There is no quality of service provisions in a typical ad hoc routing like AODV, DSR etc. To facilitate Quality of service in this project I propose Machine learning based optimal path selection to enhance QoS
in the MANET. For large quantity of data transfers needed such as video streaming, live and interactive applications or live military communications within a MANET requires long time. But as the nodes are mobile and links are prone to failure, the routing parameters also varies over time. The optimum route may not remain optimum over time. So, the problem is not only to find safe and optimum routes but to also predict when the route selection algorithm may need to be run again to react and adapt while transmitting large data.
- CSE Project Topic/Ideas 6 : Phishing attack detection based on NLP
Detection of an attack from a URL can be a trivial task even for many experienced person, because an attacker can even deceive knowledgeable users by using different techniques. The Simplest one is the use of rules However, rule
based techniques are not satisfactory For this reason, it is important to detect these attacks with software support and a single approach cannot be sufficient for a good mechanism, some hybrid approaches can be preferred.
Domain names used in punctuation attacks were tried to be detected by using Natural Language
Processing (NLP) techniques.
we have found some supporting features for the detection of such phishing attacks and evaluated the effect on the performance of the proper use of these features. Specified features alone may not be enough to check whether a domain name is malicious , but it will provide very important information in order to make a right decision. The results obtained for this purpose should be evaluated as a result of a decision support system and should be evaluated as a parameter in the final decision.
- CSE Project Topic/Ideas 7: Predicting Tags for Stack Overflow Questions Using Different Classifiers
Stack Overflow is the largest, most useful online community for developers to learn, exchange their programming skills and knowledge, build their careers and help others as well.Stack Overflow is something that every programmer and developer uses for any doubt clarification or assist others with their opinion.Every month millions of developers visit the Stack Overflow to learn, share their knowledge, and build their careers also help others. The main features of Stack Overflow are questions and answers on a wide range of topics in computer programming. The website serves as a platform for users to ask and answer questions, and through membership and active participation, to vote
- CSE Project Topic/Ideas 8 :Student Job Role Recommendation using Deep Learning
The problem statement is ‘To analyse the Job-Role for Computer Engineering Students by taking certain parameters into consideration for e.g.(Academic Interests, Coding Skills, Hackathons, Personality etc.)using Artificial Neural Networks’. For career recommendation, the feature set is very large which is to betaken into consideration, which becomes quite difficult to predict using traditional regression models. In recent years, recommendation systems have been widely used in various commercial platforms to provide recommendations for users.
A recommendation system in this domain of career prediction will thus be helpful to both, students and recruiters for choosing prospective employees. The application will help an undergraduate to choose an ideal Job-Role suited for his profile. This is possible using Artificial Neural Networking due to the advancements in the field of Deep Learning. Moreover, it will help an undergraduate to build his resume and to prepare for those specific job roles in a precise manner with
respect to gaining skill-sets and improving performance. In this project, a job role prediction system using neural networks is proposed, due to the high number of parameters for classification. These parameters include student performance in various subjects present in the undergraduate curriculum of computer science as well as student interests, interpersonal skills, talents, etc.
- CSE Project Topic/Ideas 9 : Network Monitoring using Packet Sniffer
To detect malicious users within the network or from outside the network by analyzing network traffic. Various types of attacks in the network, such as MAC Flooding, DNS Poisoning, ARP Poisoning, Dynamic Host Configuration Protocol (DHCP) Attacks, and Password sniffing. These types of attacks can be mitigated by monitoring of network in a sophisticated way. Intrusion detection can be defined as the process of observing
the behavior of the computer system or networking system for possible instruction detection by detailed analysis. An intrusion detection system can be automated by using software for detection. The automated software aims to detect the instruction and record the events for future reference. Packet sniffer accomplishes this detection.
A packet sniffer runs in a network-attached device that receives all frames of data link layer passing through the device’s network interface card. It is also known as Ethernet Sniffer or Network/Protocol Analyzer. The packet sniffer captures the packets that are addressed to other nodes, storing them for analysis. The sniffer can be used legitimately by a system or network administrator to analyze and troubleshoot the network traffic. Gathering the information captured by the packet sniffer tool, an administrator can identify malicious or erroneous packets and use the information to pinpoint bottlenecks and help maintain efficient network data transmission.
- CSE Mini Project Topic/Ideas 10 :Detection and Recovery from DoS and Eavesdropping Attacks on Wireless Networks-on-Chip
Wireless NoC are prominently prone to two major security attacks, namely eavesdropping and malicious data transfer. Protection of NoCs, whether conventional wired or newly designed Wireless NoCs, are essential against several classes of attacks, including eavesdropping and hardware trojans. Further, the added menaces that unguided wireless links impose haven’t gained much recognition. Thus, the interconnects inside Wireless NoCs are more susceptible to security attacks that led to suitable defense mechanisms. The hour’s need is to design a secure framework to counter these small yet powerful attacks on the NoCs.
This mini-project proposes a useful tool to protect NoCs from the attacks mentioned above, considering the resource limitations. The eavesdropping attack is achieved by the proposed mechanism when a malicious receiver is tuned to the same frequency and is powerful enough to amplify the signal effectively. Further, the extended mechanism prevents DoS attacks eventually caused due to the flooding of unauthenticated malicious data while a malicious transmitter is tuned to the same frequency.
Thus, the problem statement states the design of a framework that prevents the NoCs, especially wireless NoCs, from DoS and Eavesdropping Attacks which profoundly raises both security and reliability concerns in the NoCs.
- CSE Project Topic/Ideas for CSE 11 : Cross Site Scripting detection using Machine Learning
- CSE Project Topic/Ideas for CSE 12 : Hybrid Recommendation Engine Using Deep Learning
Recently, due to the outspread of contagious diseases people have started to prefer eating home cooked food for maintaining a healthy lifestyle . Also ,with the advent of technology like social media ,fitness blog posts have made people aware about having a good healthy lifestyle. But still people desire to have fast food after a while. There the
only option left to have a mouthwatering meal is to step outside or order some food from restaurants which comes at the cost of unhealthy components .
So for that there was the need to have some service industry where people could get service from chefs on demand who will make their meals in their houses so as to maintain food sanity and
proper nutrition value. Now here comes many challenges as every person’s food culture,taste,cuisine changes with mother tongue ,race and area in which they are living. Person should get the recommendation based on his mood as well as the regular cuisine he prefers. So there was the need to have a recommendation engine which could recommend the person with the appropriate chefs which he may like or prefer .
- CSE Project Topic/Ideas for CSE 13: Robust Physical world attacks in Deep Learning Visual Classification
This project is about implementing the algorithm described in the paper titled Robust Physical world attacks in
Deep Learning Visual Classification. Recent work, however, has demonstrated that DNNs are vulnerable to
adversarial perturbations . These carefully crafted modifications to the (visual) input of DNNs can cause the systems they control to misbehave in unexpected and potentially dangerous ways.
This threat has gained recent attention, and work in computer vision has made great progress in understanding the space of adversarial examples, beginning in the digital domain (e.g. by modifying images corresponding to a scene) , and more recently in the physical domain . Along similar lines, our work contributes to the understanding of adversarial examples when perturbations are physically added to the objects themselves. We choose road sign classification as our target domain for several reasons:
(1)The relative visual simplicity of road signs make it challenging to hide perturbations.
(2) Road signs exist in a noisy unconstrained environment with changing physical conditions such as the distance and angle of the viewing camera, implying that physical adversarial perturbations should be robust against considerable environmental instability.
(3) Road signs play an important role in transportation safety.
(4) A reasonable threat model for transportation is that an attacker might not have control over a vehicle’s systems, butis able to modify the objects in the physical world that a vehicle might depend on to make crucial safety
- CSE Project Topic/Ideas for CSE 14: Evaluation of Random Exponential Marking algorithm using ns-3
Mainly queuing delay is the major cause of affecting the overall performance of the network and this is introduced because of the bufferbloat problem. Internet routers are equipped with buffers but due to reduced memory costs nowadays, routers are provisioned with large-sized buffers assuming that dropping packets should be avoided
at all costs. Now because of large-sized buffers, there will be no packet drops, and identifying whether the congestion has started in the network is difficult. This breaks TCP’s congestion avoidance mechanism. The problem of bufferbloat renewed the interest in studying the Active Queue Management (AQM) algorithm and we will be
dealing mainly with the REM algorithm to address this issue.
1. Random Early Detection (RED) is an algorithm that is being implemented before REM and is different from REM because RED uses average queue length whereas REM uses a different approach to measure congestion,secondly, REM calculates the mark (or drop) probability differently than RED.
2. REM model is previously being implemented in ns-2 which is developed by the authors of REM.
3. Some other works similar to REM include implementation and evaluation of the PIE algorithm in ns-3 and its results are being compared with the model obtained from the ns-2 model of PIE.
4. Recently, the PI2 algorithm has also been implemented in ns-3, and since this algorithm is expected to perform similar to PIE, results obtained from the PI2 model of ns-3 are compared to those obtained from the PIE model of ns-3.
HOW DO YOU MAKE A MINI PROJECT IN CSE ?
- First based on you interest you should select the topic i.e topic selection.
- Research or search for the topic and read some papers related to that topic and observe if the topic can be explored more based on papers future work.
- After reading papers you should then explore the future works that is mentioned in those papers .
- Take some expert advice for starting that selected topic
- planning of work
- Finally execution.
How Do you Present Mini project ?
- Cover page and Title page
- Bonafide Certificate
- Table of Contents
- List of Tables
- List of Figures
- List of Symbols, Abbreviations and Nomenclature
Note:Take suggestions from you seniors
SUMMARY OF CSE MINI PROJECT IDEAS POST:
Above we gave you some list of project topic/ideas of cse hope this help you to choose your project.If you are not still happy with the list you can post that in the comment we will try to give you some more list of project and ideas on deep learning, machine learning and networking .And if you are looking for interview question do check our Interview section for that.