46th BCS Technical Written (ICT-281)প্রশ্ন ও সমাধান
- Set by
- BCS
- Questions
- 10 (0 MCQ, 10 written)
- Exam guide
- BCS (ICT and CSE cadre) · বিসিএস (আইসিটি ও সিএসই ক্যাডার)
- Cloud & Emerging Tech · ক্লাউড ও উদীয়মান প্রযুক্তি 1
- Computer Architecture · কম্পিউটার আর্কিটেকচার 1
- Cyber Security · সাইবার নিরাপত্তা 1
- Database · ডেটাবেজ 1
- Digital Logic · ডিজিটাল লজিক 1
- Mathematics · গণিত 1
- Networking · নেটওয়ার্কিং 1
- Operating Systems · অপারেটিং সিস্টেম 1
- Programming · প্রোগ্রামিং 1
- Software Engineering · সফটওয়্যার ইঞ্জিনিয়ারিং 1
(a) Explain the photoelectric effect and its significance in quantum theory.
Sign up free to see the model answer · উত্তর দেখতে বিনামূল্যে সাইন আপ করুন
(a) State and prove De Morgan's theorem using Boolean algebra.
Sign up free to see the model answer · উত্তর দেখতে বিনামূল্যে সাইন আপ করুন
(a) Explain the concept of pipelining and its impact on CPU performance.
Sign up free to see the model answer · উত্তর দেখতে বিনামূল্যে সাইন আপ করুন
(a) Discuss the advantages and disadvantages of Time-Division Multiple Access (TDMA) and Code-Division Multiple Access (CDMA).
Sign up free to see the model answer · উত্তর দেখতে বিনামূল্যে সাইন আপ করুন
(a) What is cryptography? Explain the working principle of RSA algorithm.
Sign up free to see the model answer · উত্তর দেখতে বিনামূল্যে সাইন আপ করুন
(a) Illustrate the difference between 'Call-by value' and 'Call-by-reference' in a function with examples.
Sign up free to see the model answer · উত্তর দেখতে বিনামূল্যে সাইন আপ করুন
(a) Distinguish between White Box and Black Box testing.
Sign up free to see the model answer · উত্তর দেখতে বিনামূল্যে সাইন আপ করুন
Differentiate between stored procedure and trigger in Structural Query Language (SQL).
Sign up free to see the model answer · উত্তর দেখতে বিনামূল্যে সাইন আপ করুন
(a) Discuss the causes and solutions of deadlock in operating systems.
Sign up free to see the model answer · উত্তর দেখতে বিনামূল্যে সাইন আপ করুন
(a) Differentiate between supervised and unsupervised learning.
Model answer
Supervised learning trains a model on labeled data (input-output pairs) so it learns a mapping from inputs to known outputs. Unsupervised learning trains on unlabeled data and finds hidden structure or groupings on its own.
| Aspect | Supervised Learning | Unsupervised Learning |
|---|---|---|
| Training data | Labeled (input + correct output) | Unlabeled (input only) |
| Goal | Learn a mapping f(x) → y to predict labels | Discover structure, patterns, or groupings |
| Feedback | Explicit error signal (loss vs true label) | No explicit error signal |
| Typical tasks | Classification, regression | Clustering, dimensionality reduction, association |
| Example algorithms | Linear/Logistic Regression, Decision Trees, SVM, Neural Networks | K-Means, Hierarchical Clustering, PCA, Apriori |
| Example use-case | Spam vs not-spam email classification | Customer segmentation from purchase history |
| Evaluation | Accuracy, precision/recall, RMSE against known labels | Silhouette score, inertia — no ground truth to compare |
Exam takeaway: Supervised = "learn from an answer key"; Unsupervised = "find the pattern with no answer key." Semi-supervised and reinforcement learning are related but distinct paradigms.
Question text is from the public recruitment exam; answers and explanations are written by ARBS Academy. All past papers · BCS (ICT and CSE cadre) guide