Question

Project Name: The keyword frequency vector (KFV). System Description: The keyword frequency vector (KFV) of a text file is a sequence of pairs of keywords and their frequency of appearance in the text. It is a good representation of the contents of the text. Keyword frequency vectors are widely used in information retrieval. For example the following is the keyword frequency vector of this paragraph. Output Example of KFV Frequency Input Word keyword frequency Ale The keyword frequency vector of a text file is a sequence of pairs of keywords and their frequency of appearance in the text. It is a good representation of the contents of the text. Keyword frequency vectors are widely used in information retrieval. For example the following is the keyword frequency vector of this paragraph. The vector clearly indicates that the paragraph is about keyword, frequency, text, and vector. When a user Searches for information about 'keyword frequency vector, this paragraph matches the query because the searched words have a high frequency of occurrences in the text. Another advantage of keyword frequency vector is that it is usually more compressive than the original text. This technique works better when small words (such as a 'the, 'is', 'it') are removed from the vector as we did in the example above. Moreover, the same word of different forms should be treated as one. For example, "keywords and "keyword' are treated as one word rather than two words in the above example. Most importantly, keyword frequency vectors can be extracted from texts automatically.

Our design problem is to design the architecture of such a system. The designer proposed the 2 following solutions. And then, you will be required to compare the advantages and disadvantages of the proposed designs based on the quality attributes

Answer 1

The design quality attributes are as follows:-

Correctness, Reliability, Adequacy, Learnability, Robustness, Maintainability, Readability, Extensibility, Testability, Efficiency, Portability

Whichever design adheres more to these principles will be a better design

So let us look at it design by design

#Design 1

This design uses a mater slave architecture where the master control coordinates over all the elemnts occuring in the text

Hence better reliabilty and robustness as all the things are things are centralized and any faults will be communicated to all segments which can then accordingly balance the load.

However in high traffic the control system will be in a bottleneck as it has to oversee all the differnet aspects and then managing, leading to potential delays in the system. Also extensions are not easy to integrate as a module has to be first created to accomodate that first in the control system.

#Design 2

The design uses a streamline architecture as the output of one segment is the input of the corresponding segment.

It has better readibility and extensibility as the stream is defined and new segments can be very easily introduced at the end of the pipeline.

Also it does dip performance during high traffic as it is continuously streaming to the next segment and not creating any buffers in between.

Hence considering these points design 2 is a better alternative than design 1.