Urdu Sentiment Analysis
Sentiment analysis(SA) is fastest growing field of Natural Language Processing, text mining under the umbrella of Artificial Intelligence more specifically Machine Learning. SA is the process of determining the polarity of sentence, word or phrase by classifying it into either as positive, negative
2025-06-28 16:36:31 - Adil Khan
Urdu Sentiment Analysis
Project Area of Specialization Artificial IntelligenceProject SummarySentiment analysis(SA) is fastest growing field of Natural Language Processing, text mining under the umbrella of Artificial Intelligence more specifically Machine Learning. SA is the process of determining the polarity of sentence, word or phrase by classifying it into either as positive, negative or neutral. Basic task is emotion recognition (affective computing & sentiment analysis). Sentiment analysis focuses on binary classification with output positive versus negative while affective computing focuses on extracting a set of emotion labels. Easy to do when emotions are expressed explicitly, difficulty arises when opinions and sentiments expressed implicitly. A text can be categorized as objective or subjective. The objective text is fact based neutral text, but opinions, reviews and discussions are all in the category of subjective text, which exhibit some feeling or sentiment. The main goal of a sentiment analyzer is to classify the subjectivity orientation towards positive or negative. That is why sentiment analysis is sometimes referred to as subjectivity analysis. Sentiment analysis can be performed at three different levels include: document level, sentence level and aspect level.
Project ObjectivesThe aim of this project is to develop an application that can identify the emotions and can mark them as either positive, negative or neutral based on the scores assign to it.
Project Implementation MethodInitially a translator will used for convering englist text into urdu after that the sentence will be broken into chunks and then polarity to each word will be assign by using sentiment analyzer and at the ens overall score will be calculated to determine the final result.
Benefits of the ProjectSentiment analysis have previously been performed by using English based lexicons by translating them to Urdu which involves extra effort as very few Urdu lexicons are available, by buliding this project this extra step can be eiminated. Tools available for sentiment analysis are limited to a polarity evaluation or a mood classification covering few sets of emotions, which can be improved by this. Additionally available these methods only works when opinions and sentiments are expressed explicitly in text, failed to identify implicit opinions and sentiments,this project focuses to overcome this limitation.
Technical Details of Final DeliverableIt will take reviews from multiple websites will convert them into chunks and assign the polarity to each word by checking its value from repositry and sum up the final polarity to identify either it is a positive opinion, negative or neutral opinion.
Final Deliverable of the Project Software SystemType of Industry IT Technologies Artificial Intelligence(AI)Sustainable Development Goals Quality Education, Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 51600 | |||
| Building web applications in PHP | Miscellaneous | 1 | 5000 | 5000 |
| python for data science | Miscellaneous | 1 | 5000 | 5000 |
| Printing | Equipment | 8 | 200 | 1600 |
| Data Bricks Amazon Cloud Server | Equipment | 1 | 40000 | 40000 |