Prevent creation of duplicate rows in database dynamically

I am new to databases so please forgive my mistakes. Here's the scenario. I use the web for my data source. Sometimes, like, for a keyword, what my crawler do is crawl same keyword repeatedly (not the crawler's fault). I don't want to include that particular keyword again in my database. I know of primary key and also know that no two primary keys can be same. So should I make my keyword primary key to avoid duplicate rows or is there some other methods to deal with it?? The DB is dynamic because there is very little manual interference in the crawl. Note: I am using Apache Cassandra.