First of all there is an important distinction to be drawn between merely a dangerous battle, and a kamikaze attack. It is understood that most operations have a risk of injury or death for soldiers but it's rare that any commander sends troops on a mission that they know will be a failure. A dangerous, unlucky or poorly planned engagement is not kamikaze attack because the latter ends with the death of a soldier even if everything goes according to plan. Events like the kamikaze pilots, suicide bombers, or mass suicides by Japanese after battles stand out because they are anomalies, even within the conduct of the Japanese military in the 20th century.

Second, while I cannot speak to Tarawa, I would not call D-Day landings suicidal by any stretch. Dangerous yes, war often is, but they were successful and despite their difficulty they were not especially bloody. Roughly 156,000 allied soldiers took part in D-day landings, they suffered a bit more than 10,000 casualties, around 6.4%, of those 4,414 died, putting the death rate at 2.8%. That is not an outstandingly bloody battle either by the number or ratio of deaths by World War 2 standards. And again, they were not sent to die, the deadliest sector: Omaha Beach was caused by a mix of heavy German defenses, and poor weather that pushed landing craft off course, delayed support, and flooded amphibious tanks. But all the landings still succeeded. A far cry from the Kamikaze attacks, which were desperation moves, with limited strategic significance.

This question is in regard to our current interpretation of past events, especially the way you place the context of your question.

Kamikaze were special attack units of the Empire of Japan. According to this definition, only Japanese soldiers (and of these particular units) qualify as such.

In the general concept of a suicide attack, the main difference with the soldiers of the boats on, say D-Day, is that the Kamikaze pilots made a conscious decision to attack and kill themselves to accomplish their purpose, while the soldiers of the Allies during the Normandy Landings (and in so many other cases in the history of the world), were taken there and most/all of them were hoping to survive, even if they knew there was a high probability to die.

While the military officers and generals who orchestrated the attack would have been aware and would have calculated the expected range of the casualties, the soldiers themselves did not make any conscious decision to end their life during the upcoming battle.

Even Leonidas' 300 Spartans (actually along with Thespians, Phocians, Helots and others, practically 5,000 to 10,000 warriors) when they fought the Persian army in Thermopylae may have not been aware that they would eventually die, even if they were willing to fight to the death, even if they understood that this was unavoidable. Therefore, they wouldn't qualify as Kamikaze.

I hope this explains what differentiates a Kamikaze attack from all other cases of practically suicide missions.