NLP: Negation and intensifiers

Make a lexicon scorer understand "not good" and "very bad".

What to build

A program that scores sentences with a lexicon, intensifiers and negation.

Requirements

  • Clean words as in the easy exercise (lowercase, strip . , ! ?). For each lexicon word at position i: if the word just before it is a booster, multiply its score by the booster's factor; then, if any negation word is among the three words before it, flip the sign.
  • For each sentence print the total with one decimal place and positive, negative or neutral.

Input

Line 1: k, then k lines word score. Next line: the negation words. Next: b, then b lines word factor. Then m and m sentences.

Output

m lines.

Constraints

k ≤ 1,000; m ≤ 100.

Example

Input:

3
good 2
bad -2
boring -2
not never
2
very 1.5
slightly 0.5
3
Not good at all.
The plot was never boring
A very bad, slightly boring film.

Output:

-2.0 negative
2.0 positive
-4.0 negative

Why it works

A booster changes how strong the next sentiment word is, and a negation shortly before a sentiment word reverses it, which fixes the most common errors of a plain word sum.

Examples

Example 1 · Input
3
good 2
bad -2
boring -2
not never
2
very 1.5
slightly 0.5
3
Not good at all.
The plot was never boring
A very bad, slightly boring film.
Expected output
-2.0 negative
2.0 positive
-4.0 negative
Example 2 · Input
1
like 2
not
1
really 2
2
I do not really like it
I really like it
Expected output
-4.0 negative
4.0 positive

Prompts inside input() are ignored when checking, so input("n: ") is fine.

Hints

Hint 1

For the word at position i, the booster is `words[i - 1]` (when i > 0) and the negation window is `words[max(0, i - 3):i]`.

Hint 2

Apply the booster factor first, then flip the sign if a negation word is in the window.

Solution

Show one correct solution (try it yourself first)
lexicon = {}
for _ in range(int(input())):
    word, value = input().split()
    lexicon[word] = float(value)
negations = input().split()
boosters = {}
for _ in range(int(input())):
    word, factor = input().split()
    boosters[word] = float(factor)
for _ in range(int(input())):
    words = [w.strip(".,!?").lower() for w in input().split()]
    score = 0.0
    for i, w in enumerate(words):
        if w not in lexicon:
            continue
        value = lexicon[w]
        if i > 0 and words[i - 1] in boosters:
            value *= boosters[words[i - 1]]
        if any(n in words[max(0, i - 3):i] for n in negations):
            value = -value
        score += value
    label = "positive" if score > 0 else "negative" if score < 0 else "neutral"
    print(f"{score:.1f} {label}")

Lesson: Sentiment analysis →