While it's true that we can build an 80,000 ton displacement aircraft carrier for about $5 billion, that's only part of the story. The price doesn't include the cost of creating the infrastructure needed to build the ship. For example, that price doesn't include the cost of building the steel mills, the specialized factories like the ones that produce nuclear powerplants and steam turbines, the shipyard itself, and other stuff needed to build the carrier like computer systems, etc.. The carrier is built in drydock and then floated. Fortunately, we don't have to spend tremendous amounts of energy lifting the thing.

Building a space station on the ground is possible because all of the infrastructure (factories, workers, suppliers) are available on the ground. Unfortunately, you're then faced with the high cost and risk of launching the thing. A Proton booster can put about 20 metric tons into LEO for about $60 million. There are other boosters under development that will be able to lift even heavier payloads but at a higher price. At Proton prices, launching 1,000 tons would cost about $3 billion. Actually, that price is low because it doesn't include a lot of other things you'd need to make it work, such as attitute control and propulsion systems on each of the modules (Protons are unmanned, so the modules have to fend for themselves).

Imagine what it would take to create similar infrastructure in space. You might be able to harvest materials from the moon or an NEO, but what would it take to create high purity alloys, then machine them to aerospace tolerances of better than 1/1000th an inch, produce electronics including fabricating computer chips, etc. Once created, this infrastructure would make it possible to build subsequent systems a lot cheaper, but the cost of creating the first one is enormous.