Lab Report

Assignment Description:

Together, your group will create a lab report using the research proposal. Each person will work on one part of the report and also help check and fix the whole thing.

First, you’ll do the experiment from your proposal to see if your ideas are right or wrong. You can follow the steps on sciencebuddies.com exactly or change them a bit based on your new information. Either way, make sure to take lots of notes or even film the experiment so you can explain it later.

Lab Report:

Optimizing Search Speed: A Computational Study

Fibonacci Group

The City College Of New York (CCNY)

Project Dates: 3/28/2024 – 4/18/2024

Abstract:

This research aims to show which type of combination of search terms will lead to the desired results: an algorithm has played a role during a search. The method that will be used for this research is to test using different search terms in Google using the “AND” and “OR” operators. Switch the order of them, record data with the amount of the results that are generated from the search engine with its required time for each attempt in different accounts, and evaluate them at the end by computing its speed. It is expected that entering specific search terms will provide fewer results, which will be the direct and desired hits and less time to be spent on generating these results, and vice versa. Additionally, less specific terms will have a higher speed (hits per second) than specific terms, which indicates more hits generated per second. These results will give the users a better searching understanding when needing a query to answer from the internet by using more specific terms and “AND”, which will be more efficient in finding the relevant information.

Introduction:

Search engines have developed into essential tools in the current digital era, acting as people’s main way of accessing information online. Users can locate pertinent information quickly thanks largely to these search engines’ efficacy and efficiency (Science Buddies). Our research aims to investigate the intricate dynamics of search engine performance, with a particular focus on the effects of various search word combinations on search speed and result accuracy. It is crucial to comprehend the dynamics underpinning search operations since society is depending more and more on search engines to get the information it needs (Schroeder). Our research is built upon the insights provided by earlier studies that examine quality and user experience with search engine adoption. There are still gaps in our knowledge of the connections between study findings and search inquiries (Sculley).

The primary goal of our research is to optimize search speed by determining the most effective methods for refining search queries to get quick and pertinent results. Our goal in investigating this issue is to add to the larger conversation about online users’ happiness with information availability. We predict that using a combination of particular search phrases will yield faster and more focused search results. We anticipate that by simplifying the practice of search engine optimization, users will be able to improve their entire search experience and find information more easily. This lab report is set up to offer a thorough examination of our study strategy, conclusions, and ramifications. Our goal is to enable users to develop knowledge to guide their search tactics and provide insight into the intricate dynamics of search engine performance through meticulous testing and data analysis.

Project Narrative (Methods):

Our research aimed to understand how different search keywords impact the results that the Google search engine returns. To guarantee the consistency and reliability of our investigation, we carefully stuck to a well-structured plan.

The experiment’s main piece of equipment was a regular Laptop with internet access. Using various search keywords, this computer functioned as the primary tool for conducting Google searches. During the experiment, we also used notebooks and pens to hand document our observations and conclusions. We were able to keep thorough records of the search results we found for every search query thanks to these tools.

In terms of the experiment itself, we used an organized approach divided into multiple separate phases. We started by deciding on a particular issue of interest associated with online information retrieval. Next, we determined a range of search phrases associated with the selected subject like “apple”, “apple AND fruit”, etc. The variables we wished to investigate in the experiment were these search phrases.

After determining the search terms, we used each term to do searches, carefully recorded the time on the notebook that was required to generate these results using a stopwatch, and used the time to calculate the speed at which it generated the results for different accounts. In this phase, we aimed to assess the quality and relevancy of the data that we retrieved for every search phrase. We went over elements including the order of relevant websites, the variety of information sources, and the precision of the data displayed in the search results.

Following the collection of the data related to search results, we went on to examine and contrast the results linked to various search phrases. To find patterns and trends, this analysis entailed looking at a variety of search result elements. We also considered any differences in the results depending on the exact search terms entered.

All things considered; The experiment’s consistency was made certain by our precise methods. Every stage from choosing search terms to running searches, gathering information, and analyzing the findings was clearly described. We followed this process in order to give others an understandable structure for testing our findings and conducting a similar study.

Results:

From the experiment, the search term “apple” was the control group, and others were the experimental group with a diversity of combinations using “AND” and “OR”. Tables 1, 2, 3, 4, and 5 displayed the five trials – as the sample size – for each of the accounts (Personal, School, Guest) on “apple”, “apple AND fruit”, “fruit AND apple”, “apple OR fruit”, and “fruit OR apple”, respectively. Table 6 displays the number of hits (in thousands) and speed (hits per second) for each account with the different combinations of search terms. Graph 1 displays the bar graph of each combination of search terms with its number of hits and speed.

Table 1. Five trials of time spent with 3 different accounts on the search term “apple”.

apple	Personal Account	School Account	Guest Account
Trial 1	0.35	0.37	0.3
Trial 2	0.4	0.41	0.37
Trial 3	0.3	0.38	0.37
Trial 4	0.31	0.32	0.36
Trial 5	0.33	0.35	0.29
Average	0.338	0.366	0.338

Table 2. Five trials of time spent with 3 different accounts on search terms of “apple AND fruit”.

apple AND fruit	Personal Account	School Account	Guest Account
Trial 1	0.41	0.32	0.27
Trial 2	0.31	0.32	0.27
Trial 3	0.35	0.32	0.27
Trial 4	0.35	0.32	0.27
Trial 5	0.35	0.31	0.3
Average	0.354	0.318	0.276

Table 3. Five trials of time spent with 3 different accounts on searched “fruit AND apple”.

fruit AND apple	Personal Account	School Account	Guest Account
Trial 1	0.31	0.31	0.29
Trial 2	0.32	0.32	0.26
Trial 3	0.32	0.33	0.26
Trial 4	0.32	0.33	0.26
Trial 5	0.32	0.33	0.29
Average	0.318	0.324	0.272

Table 4. Five trials of time spent with 3 different accounts on search terms of “apple OR fruit”.

apple OR fruit	Personal Account	School Account	Guest Account
Trial 1	0.33	0.38	0.33
Trial 2	0.33	0.38	0.26
Trial 3	0.29	0.38	0.26
Trial 4	0.29	0.35	0.28
Trial 5	0.29	0.35	0.28
Average	0.306	0.368	0.282

Table 5. Five trials of time spent with 3 different accounts on search terms of “fruit OR apple”.

fruit OR apple	Personal Account	School Account	Guest Account
Trial 1	0.48	0.32	0.26
Trial 2	0.48	0.32	0.26
Trial 3	0.48	0.32	0.27
Trial 4	0.48	0.32	0.27
Trial 5	0.48	0.32	0.27
Average	0.48	0.32	0.266

Table 6. The number of hits (in thousands) and speed (hits per second) for each account with the different combinations of search terms.

		Personal Account	School Account	Guest Account
apple	Number of Hits (in thousands)	15,650,000	15,650,000	15,650,000
	Speed (hits per second)	46301775.15	42759562.84	46301775.15
apple AND fruit	Number of Hits (in thousands)	1,130,000	1150000	1,130,000
	Speed (hits per second)	3192090.395	3616352.201	4094202.899
fruit AND apple	Number of Hits (in thousands)	1,030,000	1,030,000	1,030,000
	Speed (hits per second)	3238993.711	3179012.346	3786764.706
apple OR fruit	Number of Hits (in thousands)	21,740,000	21,740,000	21,740,000
	Speed (hits per second)	71045751.63	59076086.96	77092198.58
fruit OR apple	Number of Hits (in thousands)	13,930,000	13,930,000	13,930,000
	Speed (hits per second)	29020833.33	43531250	52368421.05

Graph 1. The bar graph of Table 6 with the number of hits and speed for each account on each combination of search terms.

From Tables 2 to 5, it was observed that the use of “AND” generally required less time to generate results compared to “OR,” which conversely demanded more time. From the bar graph, the speed associated with the use of “AND” was lower than that of “OR”, as well in the control group due to less results being generated with the use of “AND”. The obtained result indicated the specific search terms with “AND” generated fewer hits, and overall, less time was spent generating results from them for each account and vice versa with “OR”. This supported the thesis of the experiment. However, the computation indicated the speed of hits per second for the specific terms was fewer than the less specific terms for each account.

Discussion:

To conduct our lab research, which aimed to understand how various search terms impact Google search results, we followed an organized process. Our main tool was a Laptop connected to the CCNY wifi when conducting the experiment with the same browser used: Google Chrome. We mostly used google.com for the searches and carefully wrote down our findings and ensured that the method we used was consistent using standard tools like notebooks and pens. Also after every finding the cache was cleared so that the next result would not be influenced by the previous search terms. Additionally, the data is collected at the same time and the same location (CCNY) while conducting the experiment. However, the background programs are running while conducting the experiment, so the data might vary.

Unlike previous research by Chen, Schroeder, and Sculley, which focused on deep learning techniques namely, search engine knowledge shaping and hidden technical issues in machine learning systems, our experiment investigated the direct influence of search terms on search results. Our effort intended to fill the knowledge gap, even though this research provided important information regarding knowledge shaping, query optimisation and difficulties in machine learning systems. by looking into how search term selection affects user experiences and search engine outcomes in the real world. Owing to Shepard, D., claimed that individual variations exist in Google search results, applying modified search engine optimization based on browser history, search queries, and account types.

The results, showed that some search terms generated fewer hits and needed less time to create overall, proved the study’s hypothesis. However, it was found that the hits per second rate calculated with terms was slower. Our study aimed to contribute to a larger debate on online information access by using a planned method that was divided into different steps, such as identifying the problems, developing search terms, and assessment of search outcomes. Building on the results of previous studies, our research provides a thorough understanding of the complex mechanics of search engine performance and its consequences for user interactions with online information.

Works Cited:

Chen, W., Zhang, S., & Li, H. “Deep Learning Approaches for Query Optimization in Database Systems.” Accessed March 20, 2024, https://dblp.org/db/journals/tkde/tkde34.html

Science Buddies Staff. “Ready, Set, Search! Race to the Right Answer.” Science Buddies, 14 July 2023, https://www.sciencebuddies.org/science-fair-projects/project-ideas/CompSci_p015/computer-science/search-speed. Accessed 21 Mar. 2024.

Schroeder, Ralph. “Does Google Shape What We Know?” Prometheus, vol. 32, no. 2, 2014, pp. 145–60. JSTOR, https://doi.org/10.1080/08109028.2014.984469. Accessed 25 Mar. 2024.

Sculley, D., Holt, G., Golovin, D., Davydov, E., Phillips, T., Ebner, D., Chaudhary, V., Young, M., Crespo, J. F., & Dennison, D. (2015). “Hidden Technical Debt in Machine Learning Systems.” Advances in Neural Information Processing Systems, 28, 2503-2511. Retrieved from https://papers.nips.cc/paper/2015/file/86df7dcfd896fcaf2674f757a2463eba-Paper.pdf

Shepard, D, “Why Google search results vary”, August 20, 2021, https://www.titangrowth.com/blog/why-google-search-results-vary-and-how-changes-affect-seo/

The Libraries, “How do i narrow or broaden my search results” Accessed March 20, 2024, https://libanswers.libraries.wsu.edu/faq/345448

Lab Report Draft

Optimizing Search Speed: A Computational Study

Fibonacci Group

The City College Of New York (CCNY)

Project Dates: 3/28/2024 – 4/18/2024

Abstract:

Search engines have been used very commonly in recent society for answering peoples’ questions. This research aims to show which type of combination of search terms, an algorithm has played a role during a search, will lead to the desired results. The method that will be used for this research is to have the electronic device with an internet connection to test using different search terms in the search box of the search engine using the “AND” and “OR” operators, switch the order of them, and record data with the amount of the results that generating from the search engine with its required time for each attempt in different accounts, and evaluate them at the end with computing its speed. The obtained result is that entering specific search terms will provide fewer results, which will be the direct and desired hits and less time to be spent on generating these results, and vice versa. Additionally, less specific terms will have a high speed (hits per second). The conclusion is that learning the skills of using the search terms in the search engine will help people find an answer to their query with much more efficiency in finding the relevant information.

Introduction:

Search engines have developed into essential tools in the current digital era, acting as people’s main method of accessing a wealth of information online. Users can locate pertinent information fast thanks in large part to these search engines’ efficacy and efficiency. Our research aims to investigate the intricate dynamics of search engine performance, with a particular focus on the effects of various search word combinations on search speed and result accuracy. It is crucial to comprehend the dynamics underpinning search operations since society is depending more and more on search engines to get the information it needs. Our research is built upon the insights provided by earlier studies that examine quality and user experience in relation to search engine adoption. There are still gaps in our knowledge of the subtleties of the connections between study findings and search inquiries, though.

Project Narrative (Methods):

Our science fair research aimed to understand how different search keywords impact the results that the Google search engine returns. To guarantee the consistency and reliability of our investigation, we carefully stuck to a well-structured plan. We used a computer with internet access as our main tool to perform searches on the Google platform to carry out our experiment. In addition, we carefully wrote down our observations and conclusions throughout the experiment using traditional tools like notebooks and pens.

The experiment’s main piece of equipment was a regular desktop PC with internet access. Using various search keywords, this computer functioned as the primary tool for conducting Google searches. During the experiment, we also used notebooks and pens to hand document our observations and conclusions. We were able to keep thorough records of the search results we found for every search query thanks to these tools.

In terms of the experiment itself, we used an organized approach divided into multiple separate phases. We started by deciding on a particular issue of interest associated with online information retrieval. Next, we determined a range of search phrases associated with the selected subject. The variables we wished to investigate in the experiment were these search phrases.

After determining the search terms, we used each term to do searches carefully recorded the time that was required to generate these results, and computed its speed at the end. In this phase, we aimed to assess the quality and relevancy of the data that we retrieved for every search phrase. We went over elements including the order of relevant websites, the variety of information sources, and the precision of the data displayed in the search results.

All things considered; our experiment has in a way somewhat established the foundation for further research in this area while providing useful details about the complex mechanics of online information retrieval. We sought to further our understanding of the ways that search phrases affect user experiences and search engine results by carefully organizing and carrying out our efforts.

Results:

From the experiment, we had the search term “apple” as our control group, and others were our experimental group with a diversity of combinations using “AND” and “OR”. Table 1, 2, 3, 4, and 5 displayed the five trials – as sample size – for each of the accounts (Personal, School, Guest) on “apple”, “apple AND fruit”, “fruit AND apple”, “apple OR fruit”, and “fruit OR apple”, respectively. Table 6 displays the number of hits (in thousands) and speed (hits per second) for each account with the different combinations of search terms. Graph 1 displays the bar graph of each combination of search terms with its number of hits and speed. The collected data is unbiased due to the same time and location – (CCNY) used for the experiment.

Table 1. Five trials of time spent with 3 different accounts on the search term of “apple”.

Table 2. Five trials of time spent with 3 different accounts on search terms of “apple AND fruit”.

Table 3. Five trials of time spent with 3 different accounts on searched of “fruit AND apple”.

Table 4. Five trials of time spent with 3 different accounts on search terms of “apple OR fruit”.

Table 5. Five trials of time spent with 3 different accounts on search terms of “fruit OR apple”.

Table 6. The number of hits (in thousands) and speed (hits per second) for each account with the different combinations of search terms.

Graph 1. The bar graph of Table 6 with the number of hits and speed for each account on each combination of search terms

The obtained result supported the thesis of the experiment with specific search terms generating less number of hits, and overall less time spent on generating the results from them. However, the computation of the speed of hits per second with specific terms is less than the less specific terms.

Discussion:

We used a systematic approach to perform our science fair research, which aimed to understand how different search terms affect Google search results. Our primary instrument was a desktop PC with internet connectivity, and we used traditional tools like notebooks and pens to carefully record our results and make sure that our research approach was consistent.

Our experiment examined the direct influence of search phrases on search results, in contrast to earlier research by Chen, Schroeder, and Sculley that concentrated on deep learning techniques, which were search engine knowledge shaping, and hidden technical problems in machine learning systems, respectively. While these studies offered insightful information about knowledge shaping, query optimization, and technical difficulties in machine learning systems, our work aimed to close the knowledge gap by investigating the real-world effects of search phrase selection on user experiences and search engine results. Due to Shepard, D, he stated that the Google search results vary from person to person, which uses personalized search engine optimization with the search and browser history, and the variety of the accounts. This applies to the result section with the collected data varying from personal to school to guest, and how does this impact people’s search experience?

The idea behind the study was supported by the data, which showed that certain search keywords produced fewer hits and required less time to produce overall. Nonetheless, it was discovered that the calculation of the hits per second rate with phrases was slower. Using a planned approach that was broken down into several stages—such as the identification of problems, formulation of search terms, and assessment of search outcomes—our study sought to add to the larger conversation on online information access. Our research offers a detailed knowledge of the complex mechanics of search engine performance and its consequences for user interactions with online information, building on the findings of earlier studies. As we investigate the role that search terms have in influencing user experiences,

Works Cited: