Comprehensive Analysis of Peking University’s CAT 622 Computer Application Fundamentals Exam Questions

Just thinking about you following me makes me a bit nervous

Regarding the exam questions, for the vast majority of candidates, they are unknown or even terrifying. As for the interpretation of the exam questions and analysis of the topics, for most candidates, it may take several practice exams to gain some understanding, while for some candidates, even after several practice exams, it may still not be enough to fully understand the exam questions.

This is the current situation of graduate students preparing for exams, exploring key points, difficulties, and trends by doing past exam questions, because the self-designed questions are flexible and variable, making the inquiry into exam questions time-consuming and labor-intensive for candidates.

Little Sushi hopes to change this situation through our efforts. We first tell everyone about the exam situation, knowledge point distribution, question investigation direction, key and difficult points, and analysis through research and interpretation of past exam questions, and provide an analysis and prediction of the exam questions based on the situation of past years.

We hope that graduate students can familiarize themselves with the “conclusions” to feedback on their “review”, making their review targeted, precise, and efficient.

Each “topic analysis” article and “video explanation” is meticulously crafted, taking a long time, short as half a month, long as a month. Complete texts, supplemented with “video explanations”, the combination of the two is more effective.

This carefully crafted work is hoped to help you.

If you encounter problems and have questions about graduate school, or if you have articles/content/videos you want us to create, you can contact Little Sushi anytime (WeChat: ①lqll_1998 ②xiaozhushou_123), at the end of the article, there are group numbers for QQ graduate school groups in various majors. Due to some WeChat graduate school groups exceeding 200 people and the QR code expiring easily and attracting “advertisements”, directly adding Little Sushi on WeChat will get you into the group.

Table of Contents

1. Video Explanation of Topic Analysis

2. Exam Situation Analysis

3. Key Points and Question Number Analysis

4. Selected Question Analysis

5. Topic Trends

This exam situation analysis is based on the analysis of the 2019 and 2020 exam questions.

Let’s first discuss the question types from the two years:

The distribution of question points in the 2019 exam was relatively even, and the question types were quite flexible. The first big question was to choose 12 out of 18, with each worth 10 points. The second question was an application question about Office, and the third question involved two programming questions, which were a bit difficult but still solvable.

Due to the sudden change in the 2020 exam situation, this was the first time that CAT and Digital Art shared the same professional course, so the question types and types of questions changed slightly. A widely accepted reason is to accommodate the Digital Art students, as they started reviewing Computer Science later than CAT students.

The changes are as follows:

But there are still unchanged parts:

In summary, the distribution of key points can be summarized as follows:

The above statistics show how many questions were asked in each chapter. Next, we will analyze each question.

Next, Little Sushi will organize and analyze some past exam questions according to the textbook chapters.

You now have 2 memory blocks and 1 register, how do you swap data between these 2 memory blocks?

This question tests the principles of memory reading and memory addresses. The answer method is to first explain the principle of swapping data between memory blocks and registers, and then illustrate it with diagrams, text, or programming language as needed.

Two memory blocks and one register can exchange data through the bus. Assuming the two memory blocks are A and B, and the register is C, the value of A is uploaded to the bus, and then placed into a section of the data register C called C_A; the value of B is similarly uploaded to the bus and placed into a section of the data register called C_B; the value of C_B is uploaded to the bus and then assigned to A’s memory address; the value of C_A is also uploaded to the bus and assigned to B’s memory address.

(Optional answer) The above process can be represented in C++ as follows:

Assuming the current version of a single-core CPU has a clock frequency of 2GHz, and the internal workflow consists of 3 cycles: fetch, decode, execute. Propose possible methods to improve CPU capability for the next version.

This question tests the instruction execution process and how to improve CPU computing capability. When reviewing, one should not only look at the CPU part but also relate it to the entire chapter to answer. This question had a similar problem in 2017, where the question was “under the condition that overclocking is not allowed,” which was canceled in 2020. The answer method is to extract more knowledge points from the question as much as possible and answer them one by one.

The instruction execution process is as follows:

1. Retrieve the instruction from memory or from the instruction register and send it to the instruction decoder.

2. The instruction decoder decodes the instruction.

3. The control circuit generates corresponding signals to drive each component to complete the corresponding operations.

4. The program counter increments, automatically changing or pointing to the address of the next instruction.

Possible methods to enhance CPU capability:

1. Increase the CPU clock frequency. The higher the main clock frequency, the more calculations the CPU can perform, allowing it to execute more instructions quickly.

2. Improve the architecture, such as enhancing floating-point computing performance and adding dedicated floating-point instructions.

3. Increase the number of processor cores, i.e., develop multi-core processors.

4. Overclocking, increasing external frequency and multiplier frequency. By adjusting power, voltage, core, memory settings, and other important system values, top computer performance can be achieved, while also ensuring the replacement of heat sinks and fans, and increasing voltage.

5. Set up a floating-point unit (FPU) to speed up data computation.

6. Parallel processing of instructions (distributed circuits).

7. Increase cache capacity to reduce scheduling between cache and memory.

8. Increase memory capacity to reduce page scheduling between internal and external memory.

9. Regularly perform disk defragmentation, clean up system junk, and improve system speed by closing unnecessary programs in the task manager.

10. Increase the number of registers to improve bus speed.

11. Reasonably set the capacity of virtual memory.

Windows system provides a tree-structured file management system, making it easy to find and delete resources. Besides this, what other functions does the Windows file management system have, and explain their purposes (at least 5 points)

This question tests classic points in operating systems, specifically file management systems, requiring candidates to have a deep understanding of file systems. The answering approach is to recall the lecture notes learned during the study of file management systems and summarize.

The functions of the file management system are as follows:

1. Centralized storage and unified document sharing.

2. Permission management, allowing fine-grained permission control for users, departments, and positions, controlling users’ management, browsing, reading, editing, downloading, deleting, printing, subscribing, and other operations.

3. Full-text indexing, which can index the contents of Office, PDF, and other files, allowing for quick and accurate searches for required files from massive data.

4. Document auditing, which describes every action in the entire lifecycle of the document, including operator, action, date, and time. Through auditing tracking, you can grasp all file operations within the system.

5. Version management, associating multiple versions of documents to avoid using incorrect versions, while supporting viewing, reverting, and downloading historical versions.

6. Automatic numbering, allowing for the design of numbering rules.

Why do operating systems need multi-process capabilities? In what situations can multi-processing be harmful and should not be used?

This question tests the classic concepts of “threads” and “processes” in operating systems, requiring candidates to be familiar with the concept of multi-threading. This question has an out-of-syllabus content, namely “zombie processes” and “orphan processes.” If you really encounter them and are not familiar with them, my suggestion is to decisively give up and not linger, as you won’t know and neither will others. What we need to ensure is that all non-out-of-syllabus content must earn points.

When a program runs on an operating system, the operating system provides an illusion that it seems as if only this program is running, and the program appears to monopolize the processor, main memory, and I/O devices. This illusion is achieved through the concept of “process.”

The benefit of multi-processing is that it allows you to execute multiple tasks simultaneously, also known as “concurrent running,” where the instructions of one process and those of another interleave. In practical work, multiple programs often need to work simultaneously on the system, such as using DingTalk to send and receive work information, and using a browser to handle work matters, etc. If it is a single-process system, it would be much more difficult, while multi-processing solves this problem, making multi-tasking operations possible.

Harmful: Zombie processes and orphan processes.

Orphan Process: If a parent process exits while one or more of its child processes are still running, those child processes become orphan processes. Orphan processes will be adopted by the init process (process ID 1), which will collect their status information.

Zombie Process: A process creates a child process using fork, and if the child process exits without the parent process calling wait or waitpid to obtain the status information of the child process, the process descriptor of the child process remains in the system. This type of process is called a zombie process.

An application program needs to design different versions for different OS, while the OS needs to adapt to hardware architecture; however, “virtual machines” and “virtualization” allow operating systems like Linux to run on any hardware and achieve “cloud computing.” Explain the computer concepts in this sentence.

This question tests a comprehensive range of content from various chapters. Fortunately, it tests basic content, which requires memorization of definitions. Similar to the previous question, this question requires careful attention to the wording to identify all definitions involved in the question.

The computer concepts involved and their explanations are as follows:

1. Application Program: An application program is one of the main classifications of computer software, referring to software specifically written for a certain application purpose aimed at users, such as text processors, spreadsheets, accounting applications, browsers, media players, flight simulators, command-line games, image editors, etc.

2. OS: The operating system (OS) is system software that manages computer hardware and software resources and serves as the kernel and foundation of the computer system. The operating system handles basic tasks such as managing and configuring memory, determining the priority of system resource supply and demand, controlling input and output devices, operating networks, and managing file systems. It also provides an interface for users to interact with the system.

3. Hardware Architecture: The Von Neumann architecture can be answered. A computer consists of five parts: arithmetic logic unit, control unit, memory, input devices, and output devices.

4. Virtual Machine: A virtual machine is a complete computer system with full hardware functionality simulated through software, running in a completely isolated environment. Any work that can be done on a physical computer can also be accomplished in a virtual machine.

5. Cloud Computing: In simple terms, cloud computing refers to the provision of computing services (including servers, storage, databases, networking, software, analytics, and intelligence) over the Internet (the cloud), providing rapid innovation, elastic resources, and economies of scale. For cloud services, you typically pay only for what you use, helping to lower operating costs, run infrastructure more efficiently, and adjust service usage according to changing business demands.

WeChat or QQ, P2P software makes me happy (roughly this background information). Explain how these P2P software starts programs, reads contact lists, and sends text messages; what communication mode do video and voice use? (I think they might be asking about protocols), what problems does this communication mode have?

Analysis: Similar to the previous question, this question also requires attention to wording. Although the official textbook does not cover P2P-related content, this is a recently emerging concept mentioned in the “Computer Culture,” so it is not out-of-syllabus content.

P2P: Peer-to-peer (P2P) is a decentralized network system that relies on user groups (peers) to exchange information over the Internet. Its purpose is to reduce the nodes in network transmission to lower the risk of data loss.

The issue lies in “decentralization.” “Decentralization” is a phenomenon or structure that can only occur in systems with numerous users or nodes, where each user can connect and influence other nodes. In simple terms, everyone is a center, and everyone can connect and affect other nodes. This flattened, open-source, and egalitarian phenomenon or structure is called “decentralization.”

At the same time, “decentralization” is a typical feature of blockchain, which uses distributed storage and computing power. The rights and obligations of all network nodes are the same, and the data within the system is essentially maintained by all nodes in the network, making the blockchain independent of a central processing node, allowing for distributed storage, recording, and updating of data. Each blockchain follows a unified rule based on cryptographic algorithms rather than credit certificates, and the data update process requires user approval, thus establishing that blockchain does not need intermediaries or trusted institutions to endorse it.

However, decentralization brings about increased management costs.

Decentralized applications include:

Social Media

1) Didi Chuxing

The widely used bus service and the currently popular app “Didi Chuxing” form a stark contrast between “centralized” and “decentralized.” Users can personally perceive that when taking a bus, they need to go to the nearest bus stop (centralized point), and the route coverage may only reach a certain range around the destination. The emergence of “Didi Chuxing” has changed this pattern, allowing users to seek services from any accessible location, making the user the “center” during the time of use, while each vehicle providing the service is also a “center.”

2) Sina Weibo

We can use the Sina Weibo app anytime and anywhere to send information we want to convey to others. While transmitting information, we are a center; when receiving information, others are the center. This is an example of “centralization being weakened.”

Financial Services

1) Bitcoin

Bitcoin pioneered decentralized cryptocurrency, achieving a 300-fold increase in value within five years. The Bitcoin blockchain is a distributed database that adds Bitcoin symbols and specifies protocols to securely transfer them within the database without third-party involvement, thus forming a complete currency transmission system.

2) Ethereum

Ethereum is a digital currency of Ethereum, regarded as Bitcoin 2.0. Ethereum was created to address the scalability issues of Bitcoin. Developers can build applications using various modules on this platform. Developers need Ethereum to support the operation of their applications.

Content Distribution Platforms

1) Toutiao and Others

Toutiao uses personalized recommendation engine technology to recommend content based on users’ interests, locations, and various dimensions, including news, music, and games. The application essentially centers on user nodes.

What are the differences between selecting text in MS Word and Notepad, the more the better.

This question tests the basic knowledge of Word processing and also involves the text in the operating system’s Notepad software, comparing the differences between the two. To successfully answer this question, candidates need to have detailed observational experience and summarizing ability. The answering method is to compare the differences from various aspects.

The differences are as follows:

1. Word allows single-click to select a character, double-click to select a word, and triple-click to select a paragraph, while Notepad can only select part of the text individually;

2. Word allows cross-line selection using ctrl + mouse, while Notepad does not;

3. Selected text in Word can carry formatting when copied to the clipboard, while Notepad defaults to no formatting;

4. If you want to remove formatting from text in Word, you can copy the text to the clipboard and then paste it into Notepad, and then copy it back from Notepad.

Use Excel to analyze the sales situation of an e-commerce company in five cities across different fields such as enterprises, schools, etc.

This question tests how to organize and filter data in Excel. For candidates, skills such as summation, sorting, filtering, conditional formatting display, and pivot tables must be mastered.

Due to incomplete data recall, the key points are listed here:

1. To analyze sales conditions, the summation function needs to be used: Select the data to be summed, then click “AutoSum” in the “Home” tab;

2. Creating charts: Select data → Insert → Chart. Sample result:

3. Pivot Tables: Insert → Pivot Table → Select data range

Then drag the required columns, rows, and values to analyze.

1) Discuss the differences between wav/mp3 formats and midi format;

2) What similar video formats can you think of;

3) What storage and playback requirements do these video formats have for software/hardware?

4) Can you think of any cool applications for these videos?

This question tests knowledge of file formats and their representative significance. There is a lot to memorize for this section, which is almost always tested every year.

1. WAV is a waveform audio file format adopted by Microsoft, mainly recorded by external audio sources and converted into digital information by the sound card, stored with the .WAV extension, and restored to analog signals during playback, output through speakers. WAV files directly record the binary sampling data of real sound, usually resulting in large file sizes, primarily used for storing short audio clips.

MIDI does not store sound sampling information but records each note played by the instrument as a series of numbers, which are then synthesized by the synthesizer on the sound card according to the meanings represented by these numbers, outputting sound through speakers. Compared to WAV files, MIDI files are much more compact, with file sizes typically much smaller. In multimedia applications, WAV files are usually used for narration, while MIDI files are used for background music.

MP3 is a lossy compression of MPEG audio files, with a compression ratio of 10:1 to 12:1. MP3 is widely used due to its high compression, sound quality close to CD quality, ease of production, and suitability for online sharing.

Similar formats: avi, mp4

MP4 currently does not have a clear standard for detailed description, but it can be said that it is a video format developed using MPEG-4 encoding technology. Its main advantages are good quality and compression capabilities, allowing MPEG4 videos to save memory space on mobile devices while simultaneously preventing quality degradation due to excessive compression.

AVI has been around since the WIN3.1 era and is known for its good compatibility and ease of use, as well as good image quality, often being referred to alongside DVDs. However, its significant drawback is its large file size. This is also why we have seen the emergence of MPEG-1 and MPEG-4.

2. Storage: Sufficient hard drive space is required

Playback: The supporting graphics card, sound card, etc., must be complete

Software: A good player software is crucial

Hardware: Headphones, speakers, displays, etc.

3. QQ Video, Windows Media Player, etc.

Briefly describe the differences between images and graphics in terms of file formats, as well as their advantages and disadvantages.

This question tests knowledge from the multimedia chapter, specifically regarding the “differences between graphics and images.” This knowledge point is summarized in the textbook and “Computer Culture,” and understanding and memorizing it is essential. Additionally, the question also addresses image and graphic file formats, requiring candidates to remember the file formats associated with different extensions listed in the textbook.

Definition

• Images are digital images captured by input devices such as digital cameras, scanners, and camcorders, consisting of pixel arrays; image formats include png, bmp, jpg, jpeg, gif. Bitmaps are collections of pixels, also known as raster images, generally used for image processing of photographic quality, composed of many small square-like pixels. They are represented by the position and color values of the pixels, allowing for the representation of color shading changes.

• Graphics refer to vector images constructed by external contour lines, i.e., straight lines, circles, rectangles, curves, charts, etc., drawn by computers.

Data Description

• Image: Describes pixel points, intensity, and color using digital data. The storage size of the description information file is relatively large, and the object being described may lose detail or produce jagged edges when scaled.

• Graphics: Uses a set of instructions to describe the content of the graphic, such as the position, dimension, and shape of the various graphic elements. The described object can be scaled without distortion.

Screen Display

• Image: Presents the object in a digital format, displaying each point’s information at a certain resolution, allowing for direct and rapid display on the screen.

• Graphics: Uses specialized software to convert the graphic description instructions into shapes and colors on the screen.

Applicable Scenarios

• Image: Used to represent objects with a lot of detail (such as shading variations, complex scenes, and rich outline colors), such as photographs and drawings. Complex image processing is possible through image software to achieve clearer images or create special effects.

• Graphics: Describes objects with less complex outlines and less rich colors, such as geometric shapes, engineering drawings, CAD, and 3D modeling software.

Editing and Processing

• Image: Edited and processed using image processing software (such as Photoshop), mainly for conventional processing and editing of bitmap files and corresponding palette files, but cannot control transformations of specific parts. Due to the large storage space occupied by bitmaps, data compression is generally necessary.

• Graphics: Typically edited using drawing programs, producing vector graphics, allowing independent movement, scaling, rotation, and distortion of vector graphics. The main parameters are the instructions and parameters describing the position, dimension, and shape of the graphic elements.

Image Formats and Their Characteristics

Definition: Image formats refer to the methods used to store image files in computers, representing different image information—whether vector graphics or bitmap images, color numbers, and compression levels. Below, we introduce several common image and graphic file formats and their characteristics:

• PSD format: A proprietary image file format of Photoshop that supports all image types within Photoshop. However, PSD format image files are rarely supported by other software and tools. Therefore, after completing image production, it is usually necessary to convert to more common image formats for output to other software for further editing;

• BMP format: The standard Windows image format for DOS and Windows-compatible computer systems;

• JPEG format: A lossy image compression format. JPEG format loses some data information. Parsing speed is slow, image quality is low, does not support alpha channels, and the compression ratio can reach up to 1:32;

• TIFF format: A widely used bitmap image format;

• GIF format: GIF format can greatly save storage space, thus often used to save image files for web data transmission;

• EPS format: EPS format can be used to store vector graphics. When opening an EPS file containing vector graphics created by other applications in Photoshop, Photoshop rasterizes this file, converting the vector graphics into bitmap images;

• SWF format: Very suitable for delivery over the Internet because its files are small. This is because it heavily uses vector graphics. Compared to bitmap graphics, vector graphics require significantly less memory and storage space since they are represented in mathematical formulas rather than large datasets. Bitmap graphics are larger because each pixel in the image requires a separate piece of data for display.

• PNG format: Fast parsing speed, high image quality, supports alpha channels, low compression ratio;

• PNG8 format: Slow parsing speed, medium image quality, supports alpha channels, compression ratio can reach up to 1:6.

Comparison of Bitmap (Image) and Vector Graphics (Graphics)

Current electronic invoices often use digital signatures to verify authenticity. Explain how invoices are signed and verified using asymmetric encryption technology.

This question tests the application of public and private keys, requiring candidates to understand it thoroughly.

Unlike symmetric encryption algorithms, asymmetric encryption algorithms require two keys: a public key and a private key. The public key and private key are a pair; if data is encrypted using the public key, only the corresponding private key can decrypt it; if data is encrypted using the private key, only the corresponding public key can decrypt it. Because encryption and decryption use two different keys, this algorithm is called an asymmetric encryption algorithm.

Process of Asymmetric Key Encryption:

• A wants to send information to B, and both A and B must generate a pair of keys for encryption and decryption.

• A’s private key is kept secret, and A’s public key is given to B; B’s private key is kept secret, and B’s public key is given to A.

• When A wants to send a message to B, A encrypts the message using B’s public key, as A knows B’s public key.

• A sends this message to B (the message has been encrypted with B’s public key).

• When B receives this message, B decrypts A’s message with their private key; any other person who receives this message cannot decrypt it because only B has B’s private key.

• Conversely, B sending a message to A follows the same process.

Principle of Electronic Signature for Invoices:

In simple terms, an electronic signature is a password technology applied to the electronic form of documents, rather than a digital image of a handwritten signature. It is essentially an electronic code that allows the recipient to easily verify the identity and signature of the sender online. It can also verify whether the original document has changed during transmission.

If someone wants to send an important document online to someone in another location, both the sender and recipient must first apply for an electronic license from a certification authority (CA, GlobalSign). This encrypted certificate includes the applicant’s public key, referred to as the “public computer password,” used for file verification.

Upon receiving the encrypted electronic document, the recipient uses the public key issued by the CA to decrypt and read the file.

A country has military communications that need to utilize civilian lines. Referencing information security theories and cryptographic principles, what measures should be taken to ensure the safe upload and download of information and confirm data sources?

This question examines the content of the “Information Security” chapter, focusing on the application of public and private keys. This part has been explained in detail in the textbook.

Public key encryption can be used, with the private key held by the person needing to obtain the information for decryption, ensuring the security of the information; additionally, a set of public and private keys can be designed, using the private key for encryption and the public key for decryption, enabling digital signatures to confirm data sources.

Programming

Calculate the greatest common divisor (GCD) of two numbers (simple text description/ pseudocode/ flowchart is fine).

Although it is not listed as a separate programming question, it still tests programming ideas. The method for calculating the GCD is elementary mathematics, and everyone needs to recall the methods learned at that time and write them into a program to score.

Although this question requires pseudocode, Python programming is quite convenient, so the code for this question is presented in Python. In formal responses, simply change to English words.

Calculate s=a+bb+ccc+……, where b=a+1, c=b+1. Two variables can be defined, a as the first number, and N as the number of addends. For example, a=8, N=3, s=8+99+101010. You can define two functions, number2string and string2number, which only need to provide definitions without writing implementation processes.

This is the last question of this year’s exam, and it is based on a modification of the previous year’s question.

This question does not involve any data structure algorithms and is a way to test CAT programming questions; however, the thinking approach has always been a key point of assessment. The suggestion for reviewing this part is to find a programming fundamentals book to read and practice; scoring on programming questions should not be an issue.

It is important to note that since 2016, the Computer Science professional course has eliminated all multiple-choice and fill-in-the-blank questions, changing to all essay questions. This is also why only big questions were recorded before 2016.

There are two types of questions: all mandatory and some optional. The current organization is as follows:

From the summary of the above table, our advice to candidates is to be prepared for all mandatory questions.

Now, from the perspective of the question setter, let’s look ahead at the trends in question setting:

As mentioned at the beginning, the 2020 graduate exam did not test the Computer Networks section. Since the Digital Art students were notified to change their professional course in September, they would not have reviewed Computer Science as thoroughly as CAT students, making 2020 a relatively special year. This year is a special year, and special years only occur in the first year; if there is a Computer Science professional course in 2021, Digital Art students will still use the same Computer Science professional course as CAT students.

Therefore, the first trend is that the Computer Networks section will definitely return in the 21 graduate exam, becoming a difficulty for the entire exam like in the 2019 graduate exam, while the increased weight of Operating Systems and Application Fundamentals will relatively decline.

The second trend is, as can be seen from the analysis of key points, Computer Application Fundamentals, Operating Systems, Office, and Multimedia file formats remain the top priorities, while the security section has fewer questions, at least one to a maximum of two.

The third trend is, based on the past two years of exam questions, there are not many out-of-syllabus parts, with only two questions having some phrasing that is out of syllabus. Therefore, our prediction is that there will definitely be out-of-syllabus parts in the future, as out-of-syllabus parts serve two purposes: to provide some leeway for students majoring in Computer Science and to test how strong everyone’s ability to make connections is. Regarding out-of-syllabus parts, Little Sushi’s advice to everyone is:

• If there is optional space, if you see out-of-syllabus parts, you can skip them;

• If there is no optional space and you completely cannot answer, then just give up. But the non-out-of-syllabus parts are definitely the majority. As long as you review diligently, you can actually answer most of the non-out-of-syllabus questions, and your final score will not be low.

Looking back at the past two years of exam questions, if you have the habit of summarizing key points, it is not difficult to find the exam questions focus on high-frequency points tested over the years, and almost all can be found in textbooks and reference materials, which will not deviate from the high-frequency points in the next year. As for what high-frequency points there are, please look at the cookies of the study community’s senior brother’s sprint lecture:

This year, we will also conduct sprint lectures, so please look forward to it. Of course, we recommend that you sign up for our Sushi Study Society’s CAT Year-round Course to help you grasp the key points.

Finally, let’s discuss a question: how to improve scores through past exam questions? Should extra questions be practiced?

Unlike the analysis suggestions for Computer Science professional courses that recommend “practicing questions,” CAT does not recommend blindly practicing questions. The recommended process is: first read through the textbook, then start practicing past exam questions to discover the patterns of the exam, with at least two rounds of past questions.If there are seniors guiding you in studying past questions, the review will be more precise and effective!

Album丨Previous Topics’ Wonderful Articles

👇Click the image for surprises