Netease Internet product pen questions (Data Mining position)

I remember a little with my impressions, but I didn't fully remember the options. I sorted them out and discussed them with you. If there are any errors, please correct them more ~

Part 1

(All multiple choice questions are optional)

1. E1, E2, E3, and E4 are added to the stack in sequence. The following possible outbound stack sequence are ()

A. E2, E3, E4, E1

B. E1, E2, E4, E3

C. E3, E2, E4, E1

D. E2, E4, E3, E1

ABCD

2. arithmetic formula X = a + B * (C-D)/E possible suffix expression ()

A. xabcde/-* + =

B. XABC-DE/* + =

C. xabcd-* E/+ =

D. XAB + CD/e-* =

Draw the Expression Tree and traverse it in the descending order (* and + can exchange left and right subtree). Only C meets

3. Non-stable sorting ()

A. bubble sort B. Merge sort C. Fast sort D. Heap Sort E. Hill sort

Bubble is stable, and exchange is not required for merging. The fast card requires non-adjacent exchange instability. The heap is a variant of the sorting and non-adjacent switching is unstable, the elements in different insert sorting processes of Hill may be exchanged and unstable. Answer: CDE.

4. For a quad-tree with N nodes, each node has four pointers pointing to sub-nodes. How many null pointers are there _______.

A: The first node has four null pointers. Each node has three null pointers. 4 + 3 * (n-1) = 3N + 1

5. A function to find the running result

int func(int i){unsigned int temp = i;temp = (temp & 0x55555555) + (temp & 0xaaaaaaaa)>>1);temp = (temp & 0x33333333) + (temp & 0xcccccccc)>>2);temp = (temp & 0x0f0f0f0f) + (temp & 0xf0f0f0f0)>>4);temp = (temp & 0xff00ff) + (temp & 0xff00ff00)>>8);temp = (temp & 0xffff) + (temp & 0xffff0000)>>16);return temp;}

Ask func (0x7f530829) How many outputs ()

A15 blob C17 d18

The function implements the number of 1 values in the binary system. Select a for a total of 15 values.

6. Differences between processes and threads ()

A. The operating system can only schedule processes and cannot schedule threads.

B. The thread can share the memory address space, and the process cannot

C. Threads can share memory data and processes cannot

D. The process can communicate with IPC, but the thread does not work.

The operating system can schedule the thread; the process can communicate through the shared memory; the process of inter-process communication IPC is not necessarily applicable to the thread (about the process thread communication, you can look at the wonder of this summary http://blog.csdn.net/wonderwander6642/article/details/8008241 ). Answer D.

7. Which of the following statements about segment-based page management is true ()

A. Each process has one field table and one page table.

B. Each process has one segment and one page.

C. Each process has a field table and a page table.

D. One page table for each process

To implement segment-and-page storage management, the system should set a segment table for each process, including the segment number, the start address of the segment and the length of the page table. Each segment has its own page table, which records the page number of each page in the segment and the physical block number stored in the primary storage. C

8. Which of the following statements about the TCP protocol is false ()

A. When the connection is closed, both sockets must go through the time_wait status.

B. For an established TCP connection, shutdown can enable the active call end to enter semi-closed

C. By default, TCP ensures that when one end crashes, the other end can automatically detect failure.

D. Only the connections of established can send and receive information, and none of them can.

The socket end that is automatically closed enters the time_wait state and lasts for 2 MSL (maximum segment lifetime maximum lifecycle;

For example, the client shutdown the socket: when the client calls shutdown, the client sends fin to the server, the server confirms ACK, and then sends PSH ack to the TCP column, the client confirms ACK, then sends the FIN flag to the client, and the client confirms ack. Unlike closesocket closed connections, shutdown closes the data transmission channel, but does not close the data receiving channel. Therefore, the active caller has recently closed the half-way connection;

For each TCP connection, TCP manages four different timers.
(1) The retransmission timer is used when you want to receive confirmation from the other end.
(2) stick to the timer to keep the window size information flowing, even if the other end closes its receiving window.
(3) The timer can detect when the other end of an idle connection will crash or restart.
(4) The 2msl timer measures the time when a connection is in the time_wait state.

Answer

9. Which of the following statements about primary key and index are false ()

A. The unique index column can be null.

B. The foreign key in one relational table must be the primary key of the other table.

C. A relational table can have only one unique index.

D. The index is only used for query operations and has little impact on insert operations.

If multiple rows in a column contain null values, you cannot create a unique index on the column. Similarly, if multiple rows in a column combination contain null values, you cannot create a unique index on multiple columns. When creating an index, these are treated as duplicate values. The value of the primary key cannot be repeated or null ).

	Primary Key	Foreign key	Index
Definition:	Uniquely identifies a record. The record must be unique and cannot be blank.	The foreign key of the table is the primary key of the other table. The foreign key can have duplicate values and can be null values.	This field does not have repeated values, but it may have a null value.
Purpose:	Used to ensure data integrity	Used to establish connections with other tables	Is to increase the speed of query sorting
Quantity:	Only one primary key can exist.	A table can have multiple foreign keys.	A table can have multiple unique indexes.

C is incorrect.

10. There are four levels of transaction isolation in the database, of which "non-repeated read allowed" are ()

A. serializable

B. Read committed

C. Read uncommitted

D. Repeatable read

The isolation level of transactions, from the level to the high:

Read uncommitted Phantom reads, non-repeated reads, and dirty reads are allowed. 111
Read committed allows the phantom to be read and cannot be read repeatedly, but dirty reading is not allowed. 110
Repeatable read allows phantom reading, but does not allow repeated and dirty reading. Default InnoDB level. 100
Serializable Phantom reads, non-repeated reads, and dirty reads are not allowed. 000

For dirty reads that cannot be repeated, phantom read definitions can refer to http://blog.csdn.net/memray/article/details/8051893

Answer BC

11. If F (n) represents the nth item of the series, the definition is as follows:

F (1) = 1, F (2) = 2, F (n) = f (n-1) + f (n-2) (N> = 3)

Write a function to solve F (N) and list the operation complexity. The operation complexity must be less than O (N ^ 2 ).

A: programming implementation, as long as F (n) {f (n) = f (n-1) + f (n-2 );} this brute-force recursion is fine (complexity O (2 ^ n.

Part 2 professional knowledge (short answer)

1. data preprocessing is an important step in Data Mining. Please talk about your understanding of data preprocessing.

2. Give your understanding of Chinese word segmentation and talk about its difficulties and segmentation algorithms.

3. List and describe several well-known classification algorithms.

4. Describe K-means clustering algorithm

5. If you want to design a product recommendation system, please talk about your ideas.

6. Please talk about your understanding of viewpoint mining.

The basic questions of Netease Internet product test are still detailed. Some knowledge points cannot be answered if they are forgotten. The professional questions of data mining jobs are more specific, and those who have not learned the background of data mining may suffer losses.

You are welcome to discuss more ~