My first Ngram graph uses the wildcard search function to display the top 10 most popular words following "Princeton class of." I used the wildcard function for this search because I was interested to see what class years would be most popular. I was surprised to find that only one class year, 1879, was popular enough to even be plotted. There main explanation for why this was a notable class is that Woodrow Wilson graduated in 1879. Princeton also has a Class of 1879 Hall because of this that may have contributed to the popularity of the term.
My second Ngram graph combines the wildcard search function with the dependency search function. I used this to display the top 10 most popular nouns used to modify the word recipe. The graph suggests that of these 10 noun, family is by the far the most popular. I also found it interesting that the use of the word recipe (show in dark blue) was used to modify recipe, however, this seemed to drop off in recent years.
I chose to use the book The Great Gatsby for the second part of this lab.
I thought this word cloud was cool because Gatsby is in fact one of the words most commonly used in the book - and thus one of the "greatest" in size in the word cloud!
After seeing how commonly Gatsby was used from the word cloud, I used the term trend function to see at what points in the book it was most commonly used. Based on the graph, Gatsby is referred to most by his name about 2/3s of the way through the book.
I used the phrase counter tool to look up the number of times "green light" was said in the book. It wasn't as many as I expected.
I then used the bubblelines function to look at the points at which the phrases green, light, dock, and boat were used in the book. This function was similar to the term trend graph but I liked that I was able to look at the potential areas of overlap between phrases by seeing everything aligned.
The word challenge is marked as -1 but could be positive depending on the context. Likewise, combat is also marked as -1 but could have a positive connotation if used to describe fighting against something like a disease.
I didn't find any words that I thought were seriously wrong, but I don't think that the word "exempt" usually has a negative connotation. I also think that "ha," which is marked as 2, can be used in a mocking sense as much as if not more than it can be used positively.
The two analyzers agree that the sentence "His wife was shrill, languid, handsome, and horrible" is negative. They also agree that the sentence "'I've just heard the most amazing thing,' she whispered," is positive.
The two analyzers differ in their interpretations of the sentence "The lights grow brighter as the earth lurches away from the sun, and now the orchestra is playing yellow cocktail music, and the opera of voices pitches a key higher." Sentimood doesn't have a rating for this sentence, counting 0 positive and 0 negative words. The other sentiment analyzer, however, marked the sentence as positive "with a confidence of 100 percent." The analyzers also disagree about the sentence "It amazed him - he had never been in such a beautiful house before," which Sentimood marks as positive but the MeaningCloud analyzer marks as neutral.
Both analyzers agree that the sentence "Laughter is easier minute by minute, spilled with prodigality, tipped out at a cheerful word," is positive. This isn't necessarily correct when considering the use of the word "prodigality," however, which has a negative connotation and suggests the laughter described isn't genuine.
The translation of the first sentence of The Great Gatsby into Spanish using both Google Translate and Bing appears to be mostly accurate, however, neither translator retains the phrase "turning over in my mind," with both changing it to mean "spinning in my mind" in Spanish. Both translators were completely successful in translating the sentence "Gatsby's house was still empty when I left-the grass on his lawn had grown as long as mine."
Both Google Translate and Bing struggled to translate the sentence "No--Gatsby turned out all right in the end; it is what preyed on Gatsby, what foul dust floated in the wake of his dreams that temporarily closed out my interest in the abortive sorrows and short-winded elations of men." To be fair, this sentence uses obscure vocabulary that may not even have an exact, direct translation in Spanish. Similarly, both translators struggled to translate a sentence beginning with "Its vanished trees," with both translating "vanished" to "missing."
In general, I think the Google Translate and Bing did pretty well with translating sections of The Great Gatsby from English to Spanish. I think there might be some confusion if used in practice, but it would definitely provide a useful base that would probably only need to be corrected by a human translator to be used. I didn't see meaningful differences between different services in the sentences I tested today, but I think that there are some translators out there that are marginally better with tenses than others.
For my first experiment, I trained the machine to identify the difference between photos of red apples and photos of lemons. While the machine didn't work with one sample for each, it only two two samples for each class for the machine to work with very high confidence.
For my second experiment, I tried to train the machine to tell the difference between photos of my cat and photos of my dog. When I only trained the machine using a couple examples for each class, it consistently did not work. When I tested the machine again after giving six samples for each class, however, it was able to work. I originally thought this experiment would require hundreds of photos to be successful because there were factors that were different in the samples (like background, lighting, etc.) other than which pet I photographed. This, however, turned out not to have as big of an effect on how many samples it took to train the machine as I expected.