Sentiment Scorer

easyPROText ClassificationNLP 2: Applications

Score the sentiment of text using a simple lexicon-based approach.

Given a sentiment lexicon (mapping words to scores) and a text, compute the average sentiment score of all lexicon words found in the text.

Function arguments:

lexicon: a dictionary mapping each word to its sentiment score (float)
text: the text to score (lowercase words separated by spaces)

Output: Average sentiment score of matched words, rounded to 4 decimal places. If no lexicon words are found, return 0.0.

Example

Input

5
good 1.0
great 1.5
bad -1.0
terrible -2.0
ok 0.0
the movie was great but the ending was bad

Output

0.25

Reasoning

Inputs: lexicon assigns a score to 5 words (good $=1.0$ , great $=1.5$ , bad $=-1.0$ , terrible $=-2.0$ , ok $=0.0$ ), and text is a lowercase sentence of 9 space-separated words.
Matching: reading the sentence word by word, only great and bad have a score. The other 7 words (the, movie, was, but, ending, and the repeats) carry no sentiment and are ignored entirely: they do not count as zeros.
Averaging: with matched scores $s_1 = 1.5$ and $s_2 = -1.0$ , the result is $\frac{1}{n}\sum_i s_i = \frac{1.5 + (-1.0)}{2} = \frac{0.5}{2} = 0.25$ .
Output: 0.25, a single float rounded to 4 decimal places. A positive and a negative word partly cancel, leaving a mildly positive score.
Notice that the denominator is the number of matched words ( $2$ ), not the sentence length ( $9$ ). Dividing by 9 would give about $0.0556$ . A repeated lexicon word counts once per occurrence, and if nothing matches the answer is $0.0$ .

Constraints

Sentiment scores can be negative (negative sentiment) or positive
Only average over words found in the lexicon
If a word appears multiple times in text, count it each time
If no lexicon words found, return 0.0
Round to 4 decimal places

Learn this in

Introduction to NLPNLP Study Plan Text ClassificationNLP Study Plan

More from NLP 2: Applications

BIO Entity Span Extractoreasy Simple Entity Taggereasy Sentiment Word Countereasy Keyword Classifiereasy One-Hot Sequence Encodereasy

Concept primerPREVIEW

Background Knowledge

The Sentiment Scorer problem falls under the category of Text Classification, a fundamental task in Natural Language Processing (NLP). Text classification involves assigning a label or score to a piece of text based on its content. In this case, we're focusing on sentiment analysis, which aims to determine the emotional tone or attitude conveyed by the text. A simple approach to sentiment analysis is using a lexicon-based method, where a dictionary or lexicon maps words to their corresponding sentiment scores. This approach relies on the idea that the presence of certain words in a text can indicate its overall sentiment.

The given problem utilizes a sentiment lexicon, which is a pre-defined mapping of words to sentiment scores. This lexicon serves as the basis for scoring the sentiment of a given text. The scores are typically floating-point numbers, where positive values indicate positive sentiment, negative values indicate negative sentiment, and values close to zero indicate neutral sentiment. Understanding how to work with such a lexicon and how to compute the average sentiment score of words found in a text is crucial for solving this problem.

In the context of NLP, tokenization (the process of breaking down text into individual words or tokens) and lookup (finding the sentiment score of each word in the lexicon) are key operations. The problem also involves basic statistical computation, specifically calculating the average of the sentiment scores of matched words. Understanding these concepts and how they apply to text data will help in developing an effective solution to the Sentiment Scorer problem.

Algorithm/Approach

The general approach to solving this problem involves a straightforward lexicon-based sentiment analysis algorithm:

Load the sentiment lexicon into a data structure that allows for efficient lookup.
Tokenize the input text into individual words.
For each word in the text, check if it exists in the lexicon and if so, retrieve its sentiment score.
Calculate the average sentiment score of all words found in the lexicon. This approach is simple yet effective for demonstrating the basics of sentiment analysis using a pre-defined lexicon.

PRO

Continue the full explanation

You're reading the free preview. Unlock the complete walkthrough, the code editor, test runner and reference solution with Premium.

PRO

Editor locked

The code editor is locked for Pro problems. It is only available for free problems. Please upgrade to gain access to the code editor for all problems.

solution.py

Python 3.13

Test results