AI/ML Project

✔

In the hospitality sector, capturing and understanding customer feedback has been challenging.

✔

Feedback was collected from various sources such as online reviews, social media posts, and traditional feedback forms, leading to data centralization and analysis issues.

✔

Business owners, especially C-level executives, struggled to gauge the performance of specific business ventures due to scattered and fragmented data.

✔

Gibberish reviews cluttered the data, consuming space without adding value.

✔

There was no dedicated dashboard to measure the performance of particular business verticals, complicating business expansion and leaving over 40% of negative reviews unactionable.

✔

We implemented sentiment analysis to process customer feedback data from various sources, including online reviews, social media posts, and feedback forms. APIs were designed to capture these reviews and map them to our database.

✔

Pandas was used for preprocessing and analyzing large volumes of text data, removing noise and irrelevant information.

✔

Google BERT was employed to identify relevant keywords from reviews, classifying them as positive, negative, or neutral.

✔

CNN was trained to learn features from different parts of sentences, aiding in training the dataset.

✔

Deep learning models were used to understand complex and non-linear relationships between input features and output labels, making them ideal for sentiment analysis tasks.

✔

Based on the analysis, the hotel chain identified areas needing improvement such as slow service, outdated facilities, and unclean rooms.

✔

Investments were made in staff training, facility renovations, and increased room cleaning frequency.

✔

Positive feedback was used to highlight strengths and unique selling points in marketing campaigns.

✔

A single unified dashboard was created to capture all data and provide an overall sentiment view of a particular business within just two clicks.

✔

Responding to reviews resulted in a 12% increase in revenue per available room (RevPAR) compared to hotels that did not respond.

✔

A 12% monthly revenue increase was achieved by taking action on received feedback.

✔

Guest loyalty surged by over 10% due to responsiveness to feedback.