Remember the first time Baidu phone interview

Today, from two o'clock in the afternoon to three, the first telephone interview was conducted. Lasts 59 minutes, pure technical surface.

Straight into the subject, began the study of the algorithm. A total of two, about 40 minutes, the first in the hint reluctantly answered, the second is really not answered.

Then there are some problems related to the project, especially the information extraction module, and then asked me some special cases of some ideas and so on, alas ... There is no thought of being asked.

Before a bit of preparation is not, still busy project things, originally some algorithms do not understand, this time more tragic ...

Here are the details:

The first algorithm problem: There is an array, set it to N1, how to remove the first N2 large number. And then with the qualification, assuming that the numbers are between 0-10000.

Before I began to see this topic, I thought of the heap sort. Because before have seen, very large an array, take the first few big numbers words, with the smallest heap. Take the first n small number, then use the maximum heap. The results asked me to implement the process, no impression--. Forget about the biggest minimum heap what's going on ... Die, open this article to continue to see

The interviewer says talk about your thoughts, no matter how they come true. I said we can consider a simple sort first, then take the N2 before the OK. He asked the complexity of how much (he likes to ask the complexity, if you say the idea can be achieved, ask you the complexity, the whole process down, I have about 10 or so complexity), I said simple bubble sort, such as, should be N-side. He said there is no easy way, I say quick sort words, should be nlogn. Then let me tell the following quick sort, this seems to be barely clear--.

There was no further progress, the interviewer suggested that the condition was specific and the numbers were between 0-10000. Do you think it is possible to count 10000 how many, 9999 have how many and so on. And then...... It's glorious, I ran off again (--. I do not know what to think, all prompted to this extent, I said can binary find, 5000-10000 of how many, then not enough words, 2500-5000 how much. Ask the final complexity of how much, I forget, seemingly or nlogn. - -。

Then, after thinking about it, I gave the answer, which is the end of the first question.

When I came back to discuss with my classmates, I thought that I could sort it by base ... It's almost like the interviewer's clue.

the second algorithm problem: There are two ordered arrays N1 and N2, how to find out the number of bits.

With the first lesson, from the simple to come, say you can merge the sort first, then take the median. The interviewer asks, what is the complexity of the ordering of these two ordered arrays? I again gorgeous 2 b again. Just started to say min{n1,n2}, and then after thinking, the affirmative answer is MAX{N1,N2}. Then the interviewer asks why, then various explanations. Finally, the interviewer said, whether it is max{n1,n2} or N1+N2, we continue to look down. I suddenly want to smoke their own impulse, a n1+n2 ah--.

To keep looking down means that there is no easy way. I twisted my bile, I had no idea ... (How can you be so stupid?) Then, the interviewer again prompts: If you give you a number of median size, how to insert it. Instantly thought of binary insertion, respectively, from the middle to find. Then ask the complexity, answer Yue Logn1+logn2. And then it goes on, which is basically a hint of the way to use binary. But after a long deliberation, still did not give the final solution. Alas, helpless into the next step--.

Project Experience:

First of all: Metasearch, let me simply say. And then asked me which parts I was responsible for, and I said I did it all. It seems that people are not very interested, and went to the following one.

The second is the information extraction aspect, detailed asked my realization idea. And then raised a number of questions, and finally for some video sites, the text is particularly small, there is no idea? I twisted my bile again, saying that for large sites you can do rule-based extraction. The interview of the elder brother asked, that some of the site, or a little bit of ideas did not--.

The next time is almost there, the following text classification, Hadoop, MPI did not ask the end, and then asked me if there is anything to ask him. I asked him to combine the interview performance, give some advice to God horse. He replied that basic algorithms and data structures should be understood in depth, can not stay on the surface, to think more, have their own ideas.

Well, my first time was so gorgeous, the end of the terrible--.